Fund the Washington Area Transit System With A Mandatory Fee on Commuter Parking Spaces

A.  Introduction

The Washington region’s primary transit authority (WMATA, for Washington Metropolitan Area Transit Authority, which operates both the Metrorail system and the primary bus system in the region) desperately needs additional funding.  While there are critical issues with management and governance which also need to be resolved, everyone agrees that additional funding is a necessary, albeit not sufficient, element of any recovery program. This post will address only the funding issue.  While important, I have nothing to contribute here on the management and governance issues.

WMATA has until now been funded, aside from fares, by a complex set of financial contributions from a disparate set of political jurisdictions in the Washington metropolitan region (four counties, three municipalities, plus Washington, DC, the states of Maryland and Virginia, and the federal government, for a total of 11 separate political jurisdictions). Like for governments everywhere, budgets are limited.  Not surprisingly, the decisions on how to share out the costs of WMATA are politically difficult, and especially so as a higher contribution by one jurisdiction, if not matched by others, will lead to a lower share in the costs by those others.  And unlike most large transit systems in the US, WMATA depends entirely (aside from fares) on funding from political jurisdictions.  It has no dedicated source of tax revenues.

This is clearly not working.  Everyone agrees that additional funding is needed, and most agree that a dedicated funding source needs to be created to supplement the funds available to WMATA.  But there is no agreement on what that additional funding source should be.  There have been several proposals, including an increase in the sales tax rate in the region or a special additional tax on properties located near Metro stations, but each has difficulties and there is no consensus.  As I will discuss below, there are indeed issues with each.  They would not provide a good basis for funding transit.

The recommendation developed here is that a fee on commuter parking spaces would provide the best approach to providing the additional funding needed by the Washington region’s transit system.  This alternative has not figured prominently in the recent discussion, and it is not clear why.  It might be because of an unfounded perception that such a fee would be difficult to implement.  As discussed below, this is not the case at all.  It could be easily implemented as part of the property tax system that is used throughout the Washington region.  It should be considered as an approach to raising the funds needed, and would perhaps serve as an alternative that could break the current impasse resulting from a lack of consensus for any of the other alternatives that have been put forward thus far.

Four factors need to be considered in any assessment of possible options to fund the transit systems.  These are:

  • Feasibility:  Would it be possible to implement the option in practical terms?  If it cannot be implemented, there is no point in considering it further.
  • Effectiveness:  Would the option be able to raise the amount of funds needed, with the parameters (such as the tax rates) at reasonable levels that would not be so high as to create problems themselves?
  • Efficiency:  Would the economic incentives created by the option work in the direction one wants, or the opposite?
  • Fairness:  Would the tax or option be fair in terms of who would pay for it?  Would it be disproportionately paid for by the poor, for example?

This blog post will assess to what degree these four tests are met by each of several major options that have been proposed to provide additional funding to WMATA.  A mandatory fee on parking spaces will be considered first, and in most detail.  Many will call this a tax on parking, and that is OK.  It is just a label.  But I would suggest it should be seen as a fee on rush hour drivers, who make use of our roads and fill them up to the point of congestion.  It can be considered similar to the fees we pay on our water bills – one would be paying a fee for using our roads at the times when their capacity is strained.  But one should not get caught up in the polemics:  Whether tax or mandatory fee, they would be a charge on the parking spaces used by those commuters who drive.

Other options then considered are an increase in the bus and rail fares charged, an increase in the sales tax rate on all goods purchased in the region, and enactment of a special or additional property tax on land and development close to the Metrorail stations in the region.

No one disputes that enactment of any of these taxes or fees or higher fares will be politically difficult.  But the Washington region would collapse if its Metrorail system collapsed.  Metrorail was until recently the second busiest rail transit system in the US in terms of ridership (after New York).  However, Metrorail ridership declined in recent years, to the point that it was 17% lower in FY2016 than what it was in FY2010.  The decline is commonly attributed to a combination of relatively high fares, lack of reliability, and the increased safety concerns of recent years, combined most recently with periodic shutdowns on line segments in order to carry out urgent repairs and maintenance. Despite this, Metrorail in 2016 was still the third busiest rail system in the country (just after Chicago).

But the Washington region cannot afford this decline in transit use.  Its traffic congestion, even with Metro operating, is by various measures either the worst in the nation or one of the worst.  Furthermore, the traffic congestion is not just in or near the downtown area.  As offices have migrated to suburban centers over the last several decades, traffic during rush hour is now horrendous not simply close to the city center, but throughout the region. See, for example, this screen shot from a Google Maps image I took at typical weekday afternoon during rush hour (5:30 pm on Tuesday, April 18):

The roads shown in red have traffic backed up.  The congestion is bad not simply around downtown, nor simply on the notoriously congested Capital Beltway as well, but also on roads at the very outer reaches of the suburbs.  The problem is region-wide, and it is in the interest of everyone in the region that it be addressed.

A good and well-run transit system will be a necessary component of what will be needed to fix this, although this is just the minimum.  And for this, it will be fundamental that there be a change in approach from a short-term focus on resolving the immediate crisis by some patch, to a perspective that focuses on how best to utilize, and over time enhance, the overall transportation system assets of the Washington region.  This includes both the Metro system assets (where a value of $40 billion has been commonly cited, presumably based on its historical cost) but also the value of the highways and bridges and parking facilities of the region, with a cost and a value that would add up to far more. These assets are not well utilized now.  A proper funding system for WMATA should take this into account.  If it is not, one can end up with empty seats on transit while the roads are even more congested.

The first question, however, is how much additional funding is required for WMATA.  The next section will examine that.

B.  WMATA’s Additional Funding Needs

How much is needed in additional funding for WMATA?  There is not a simple answer, and any answer will depend not only on the time frame considered but also on what the objective is.

To start, the FY18 budget for WMATA as originally drawn up in the fall of 2016 found there to be a $290 million gap between expenditures it considered to be necessary based on the current plans, and the revenues it forecast it would receive from fares (and other revenue generating activities such as parking fees at the stations and from advertising) and what would be provided under existing formulae from the political jurisdictions.  This gap was broadly similar in magnitude to the gaps found in recent years at a similar stage in the process.  And as in earlier years, this $290 million gap was largely closed by one-off measures that one could not (or at least should not) be used again.  In particular, funds were shifted from planned expenditures to maintain or build up the capital assets of the system, to cover current operating costs instead.

Looking forward, all the estimates of the additional funding needs are far higher.  To start, an analysis by Jeffrey DeWitt, the CFO of Washington, DC, released in October 2016 as part of a Metropolitan Washington Council of Governments (COG) report, estimated that at a minimum, WMATA faced a shortfall over the next ten years averaging $212 million per year on current operations and maintenance, and $330 million per year for capital needs, for a total of $542 million a year.  This estimate was based on an assumption of a capital investment program summing to $12 billion over the ten years.

But the “10-Year Capital Needs” report issued by WMATA a short time later estimated that the 10-year capital needs of WMATA would be $17.4 billion simply to bring Metro assets up to a “state of good repair” and maintain them there.  It estimated an additional $8 billion would be needed for modest new investments – needed in part to address certain safety issues.  But even if one limited the ten-year capital program to the $17.4 billion to get assets to a state of good repair, there would be a need for an additional $540 million a year over the October 2016 DeWitt estimates, i.e. a doubling of the earlier figure to almost $1.1 billion a year.

A more recent, and conservative, figure has been provided by Paul Wiedefeld, the General Manager of WMATA, in a report released on April 19.  He recommended that while Metro has capital needs totaling $25 billion over the next ten years, he would propose that a minimum of $15.5 billion be covered for the system “to remain safe and reliable”.  Even with this reduced capital investment program, he estimated that if funding from the jurisdictions remained at historical levels, there would be a 10-year funding gap of $7.5 billion remaining.  If jurisdictional funding were to rise at 3% a year in nominal terms, then he estimated that $500 million a year would still be necessary from some new funding source.

But this was just for the capital budget, and a highly constrained one at that.  There would, in addition, be a $100 million a year gap in the operating budget, even with the funding from the jurisdictions for operations rising also at 3% a year.  Wiedefeld suggested that it might be possible to reduce operating costs by that amount.  However, this would require cutting primarily labor expenditures, as direct labor costs account for 74% of operating expenditures.  Not surprisingly, the WMATA labor union is strongly opposed.

Even more recently, the Metropolitan Washington Council of Governments issued on April 26 the final report of a panel it convened (hereafter COG Panel or COG Panel Report) that examined Metro funding options.  The panel was made up of senior local administrative and budget officials.  While the focus of the report was an examination of different funding options (and will be discussed further below), it took as a basis of its estimated needs that WMATA would need to cover a ten-year capital investment program of $15.6 billion (to reach and maintain a “state of good repair” standard).  After assuming a 3% annual increase in what the political jurisdictions would provide, it estimated the funding gap for the capital budget would sum to $6.2 billion. Assuming also a 3% annual increase in funding from the political jurisdictions for operations and maintenance (O&M), it estimated a remaining funding gap of $1.3 billion for O&M.  The total gap for both capital and O&M expenses would thus sum to $7.5 billion over the period.

But while these COG estimates were referred to as 10-year funding gaps (thus averaging $750 billion per year), the table in its PowerPoint presentation on the report on page 13 makes clear that these are actually the funding gaps for the eight year period of FY19 to FY26.  FY17 is already almost over, and the FY18 budget has already been settled.  For the eight year period from FY19 going forward, the additional funding needed averages $930 million per year.  The COG Panel recommended, however, a dedicated funding source that would generate less, at $650 million per year to start (which it assumes would be in 2019).  But the reason for this difference is that the COG Panel recommended also that WMATA borrow additional funds in the early years against that new funding stream, so as to cover together the higher figure ($930 million on average per year over FY19-26) for what is in fact needed.  While such borrowing would supplement what could be funded in the early years, the resulting debt service would then subtract from what one could fund later.  While prudent borrowing certainly has a proper role, future funding needs will certainly be higher than what they are right now, and thus this will not provide a long-term solution to the funding issue.  More funding will eventually (and soon) be required.

All these figures reviewed thus far assume capital investment programs only just suffice to bring existing assets up to a “state of good repair”, with nothing done to add to these assets.  It also appears that the estimates were influenced at least to some extent by what the analysts thought might be politically feasible.  Yet additional capacity will be needed if the Washington region is to continue to grow.  While these additional amounts are much more speculative, there is no doubt that they are large, indeed huge.

The most careful recent study of long-term expansion needs is summarized in a series of reports released by WMATA in early 2016.   A number of rail options were examined (mostly extensions of existing rail lines), with the conclusion that the highest priority for a 2040 time horizon was to enhance the capacity at the center of the system.  Portions of these lines are already strained or at full capacity, including in particular the segment for the tunnel under the Potomac from Rosslyn.  Under this plan, there would be a new circular underground loop for the Metro lines around downtown Washington and extending across the Potomac to Rosslyn and the Pentagon.  It is not clear that a good estimate has yet been done on what this would cost, but the Washington Post gave a figure of $26 billion for an earlier variant (along with certain other expenditures).  This would clearly be a multi-decade project, and if anything like it is to be done by 2040, work would need to begin within the current 10-year WMATA planning horizon.  Yet given WMATA’s current difficulties, there has been little focus on these long-term needs.  And nothing has been provided for them.

To sum up, how much in additional funding is needed?  While there is no precise number, in part because the focus has been on the immediate crisis and on what might be considered politically feasible, for the purposes of this post we will use the following.  At a minimum, we will look at what would be needed to generate $650 million per year, the same figure arrived at in the COG Panel Report.  But this figure is clearly at the low end of the range of what will be needed.  At best, it will suffice only for a few years.  Our political leaders in the region should recognize that this will need to rise to at least $1 billion per year within a few years if necessary investments are to be made to ensure the system not only reaches a “state of good repair” but also sustains it.  Furthermore, it will need to rise further to perhaps $2.0 billion a year by around 2030 if anything close to the system capacity that will be needed by 2040 is to be achieved.

For the analysis below, we will therefore look at what the rates will need to be to generate $650 million a year at the low end and roughly three times this ($2.0 billion a year in nominal terms, by the year 2030) at the high end.  These figures are of course only illustrative of what might be required.  And for the forecast figures for 2030, I will assume (consistent with what the COG Panel did) that inflation from now to then will rise at 2% a year while real growth in the region will rise, conservatively, at 1% a year.  Note that $2.0 billion in 2030 in nominal terms would be equivalent to $1.55 billion in terms of dollars of today (2017) if inflation rises at 2% a year.

It is important to recognize that providing just the low-end figure of $650 million a year will not suffice for more than a few years.  It does provide a starting point, and while that is important, when considering such a major reform as moving to a dedicated funding source to supplement government funding sources, one should really be thinking longer term.  Not much would be gained by moving to a funding source which would prove insufficient after just a few years, leading to yet another crisis.

C.  A Mandatory Fee on Commuter Parking Spaces

A fee would be assessed (generally through the property tax system) on all parking spaces used by office and other commuting employees.  It would not be assessed on residential parking, nor on customer parking linked to retail or other such commercial space, but would be limited to the all-day parking spots that commuters use.

It would be straightforward to implement.  The owners of the property with the parking spaces would be assessed a fee for each parking space provided.  For example, if the fee is set at $1 per day per space, a fee of $250 per year would be assessed (based on 250 work-days a year, of 52 weeks at 5 days per week less 10 days for holidays).  It would be paid through the regular property tax system, and collected from the owners of that land along with their regular property taxes on the semi-annual (or quarterly or whatever) basis that they pay their property taxes. The owners of the spaces would be encouraged to pass along the costs to those employees who drive and use the spaces (and owners of commercial parking lots will presumably adjust their monthly fees to reflect this), but it would be the owners of the parking spaces themselves who would be immediately liable to pay the fees.

Property records will generally have the number of parking spaces provided on those plots of land.  This will certainly be so in the cases of underground parking provided in modern office buildings and in multi-story commercial parking garages.  And I suspect there will similarly be such a record of the number of spaces in surface parking lots.  But even if not, it would be straightforward to determine their number.  Property owners could be required to declare them, subject to spot-checks and fines if they did not declare them honestly. One can now even use satellite images available on Google Maps to count such spaces. And a few years ago my water bills started to include a monthly fee for the square footage of impermeable space on my land (from roofs and driveways primarily), as drainage from such surfaces feed into stormwater drains and must ultimately be treated before being discharged into the Potomac river.  They determined through the property records system and from satellite images the square footage of such spaces on all individual properties.  If that can be done, one certainly determine the number of parking spaces on open lots.

There are, however, a few special cases where property taxes are not collected and where different arrangements will need to be made.  But this can be done.  Specifically:

  1. Properties owned by federal, state, and local governments will generally not pay property taxes.  But the mandatory fees on parking spaces could still be collected by these government entities and paid into the system just as by private property owners.  Presumably, the governments support the reform as it is supplementing the funds they already provide to WMATA.
  2. Similarly, international organizations located in the Washington region, such as the World Bank, the IMF, the Inter-American Development Bank, and others (mostly much smaller) operate under international treaties which provide that they do not owe property taxes on properties they own.  But as with governments, they could collect such fees on parking spaces made available to their employees who drive to work.  They already charge their employees monthly fees for the spaces, and the new fee could be added on.  And while I am not a lawyer, it might well be the case that such a fee on parking spots could be made mandatory.  The institutions do pay the fees charged for the water they use, and employees do pay sales taxes on the food they purchase in their cafeterias.  Finally, these institutions advise governments to apply good policy.  The same should apply here.
  3. There are also non-profit hospitals, universities, and similar institutions, which are major employers in the region but which may not be charged property taxes. However, the fee on parking spaces, while collected for most through the property tax system, can be seen as separate from regular property taxes.  It is a fee on commuters who make use of our road system and add to its congestion.  The parking fees could still be collected and paid in, even if no regular property taxes are due.
  4. Finally, the Washington region has a large number of embassies and other properties with strict internationally recognized immunities.  It might well be the case that it will not be possible to collect such a mandatory fee on parking spots for their employees (although again, presumably the embassies pay the fees on their water bills).  But the total number employed through such embassies is tiny as a share of total employment in the DC region.  And some embassies might well pay voluntarily, recognizing that they too are members of the local community, making use of the same roads.  Finally, note that embassy employees with diplomatic status also do not pay sales tax on their day-to-day purchases, while the embassy compounds themselves do not pay property taxes.  Proposals to fund WMATA through new or higher property taxes or sales taxes (discussed below) will face similar issues.  But as noted above, the amounts involved are tiny.

How, then, would such a mandatory fee on commuter parking spaces stand up under the four criteria noted above?:

a)  Feasibility:  As just discussed, such a fee on commuter parking spaces, implemented generally through the regular property tax system, would certainly be feasible.  It could be done.  It may well be that a lack of recognition of this which explains why such an option has typically not been much considered when alternatives are reviewed for how to fund a transit system such as WMATA.  It appears that most believe that it would require some system to be set up which would mandate a payment each day as commuters enter their parking lots.  But there is no need for that.  Rather, the fee could be imposed on the owner of the parking space, and collected as part of their property tax payments.  It would be up to the owner of that space to decide whether to pass along that cost to the commuters making use of those spaces (although passing along the cost should certainly be encouraged, so that the commuters face the cost of their decision to drive).

b)  Effectiveness:  The next question is whether such a fee, at reasonable rates, would generate the funds needed.  To determine this, one first needs to know how many such parking spots there are in the Washington region.  While more precise figures can be generated later, all that is needed at this point is a rough estimate.

As of January 2017, the Bureau of Labor Statistics estimated there were 3,217,400 employees in the Washington region’s Metropolitan Statistical Area (MSA).  While this MSA area is slightly larger than the jurisdictions that participate in the WMATA regional compact, the additional counties at the fringes of the region are relatively small in population and employment.  This figure on regional employment can then be coupled with the estimate from the most recent (2016) Metropolitan Washington COG “State of the Commute” survey, which concluded that 61.0% of commuters drive alone to work, while an additional 5.4% drive in either car-pools or van-pools.  Assuming an average of 2.5 riders in car-pools and van-pools (van-pools are relatively minor in number), this would work out to 63.2% as the number of cars (as a share of total employment) that carry commuters to their jobs.  Applying the 63.2% to the 3,217,400 figure for the number employed, an estimated 2,033,400 cars are used to carry commuters.  The total number of parking spaces will be somewhat more, as the parking lots will normally have some degree of excess capacity, but this can be ignored for the estimate here.  Rounding down, there are roughly 2 million parking spaces for these cars in the DC region.  And this number can be expected to grow over time.

With 2 million parking spaces, a daily fee of $1 would generate $500 million per year (based on 250 work-days per year).  A fee of $1.30 per day would generate $650 million. And assuming commuter parking spots grow at 1% a year (along with the rest of the regional economy) to 2030, a $3.50 fee in 2030 would generate $2.0 billion in the prices of that year (equivalent to $2.70 per day in the prices of 2017, assuming 2% annual inflation for the period).

Compared to the cost of driving, fees of $1.30 per day or even $3.50 per day are modest. While many workers do not pay for their parking (or for the full cost of their parking), the actual cost can be estimated by what commercial parking firms charge for their monthly parking contracts.  For the 33 parking garages listed as “downtown DC” on the Parking Panda website, the average monthly fee (showing on April 29, 2017) was a bit over $270. This would come to $13 per work day (based on 250 work days per year).  While the charges will be less in the suburbs, there will still be a cost.  But the full cost to commuters to drive to work is in fact much more.  Assuming the average cost of the cars driven is $36,000, and with simple straight line depreciation over 10 years, the average monthly cost will be $300. To this one should add the cost of car insurance (on the order of $50 to $100 per month), of expected repair costs (probably of similar magnitude), and of gas. The full cost of driving would on average then total over $600 per month, or about $29 per work day.  Even if one ignores the cost of the parking spot itself (as drivers will if their employers provide the spots for free), the cost to the driver would still average about $16 per work day.  An added $1.30 per day to cover the funding needs of the public transit system is minor compared to any of these cost estimates, and would still be modest at $3.50 per day (equal to $2.70 in the prices of today).

Thus at reasonable rates on commuter parking spots, it would be possible to collect the $650 million to $2.0 billion a year needed to help fund WMATA.

c)  Efficiency:  Another consideration when choosing how best to provide additional funds to WMATA is the impact on efficiency of that option.  A fee on parking spaces would be a positive for this.  The Washington region stands out for its severe congestion, including not only in the city center but also in the suburbs (and often even more so in the suburbs).  A fee on parking spots, if passed along to the commuters who drive, would serve as an incentive to take transit, and might have some impact on those at the margin. The impact is likely to be modest, as a $1.30 to $3.50 fee per day would not be much.  As just discussed above, given the current cost of driving (even when commuters who drive are not charged for their parking spots), an additional $1.30 to $3.50 would be only a small additional cost, even when it is passed along.  But at least it would operate in the direction one wants to alleviate traffic congestion.

d)  Fairness:  Finally, the fee would be fair relative to the other options being considered in terms of who would be impacted.  Those who drive to work (over 90% of whom drive alone) are generally of higher income.  They can afford the high cost of driving, which is high (as noted above) even in those cases when they are provided free parking spaces by their employer.

Some would argue that since the drivers are not taking transit, they should not help pay for that transit.  But that is not correct.  First of all, they have a direct interest in reducing road congestion, and only a well-functioning transit system can help with that.  Drivers benefit directly (by reduced congestion) for every would-be driver who decides instead to take transit.  Second, all the other feasible funding options being considered for WMATA will be paid for in large part by drivers as well.  This is true whether a higher sales tax is imposed on the region, higher property taxes, or just higher government funding from their budgets (with this funding coming from the income taxes as well as sales taxes and property taxes these governments receive).  And as discussed below, higher fares on WMATA passengers to raise the amounts needed is simply not a feasible option.

Some drivers will likely also argue that they have no choice but to drive.  While they would still gain by any reduction in congestion (and would lose in a big way due to extreme congestion if WMATA service collapses due to inadequate funding), it is no doubt true that at least some commuters have no alternative but to drive.  However, the number is quite modest.  The 2016 survey of commuters undertaken by the Metropolitan Washington COG, referred to also above, asked their sample of commuters whether there was either bus service or train service “near” their homes (“near” as they would themselves consider it), and separately, “near” their place of work.  The response was 89% who said there were such transit services near their homes, and 86% who said there were such transit services near their places of work.  But note also that the 11% and 14%, respectively, who did not respond that there was such nearby transit, included those who responded that they did not know.  Many of those who drive to work might not know, as they never had a need to look into it.

The share of the Washington region’s population who do not have access to transit services is therefore relatively small, probably well less than 10% of commuters.  The transit options might not be convenient, and probably take longer than driving in many if not most cases given the current service provision, but transit alternatives exist for the overwhelming share of the regional population.  The issue is that those who can afford the high cost will drive, while the poorer workers who cannot will have no choice but to take transit.  Setting a fee on parking spaces for commuters in order to support the maintenance of decent transit services in the region is socially as well as economically fair.

D.  Alternative Funding Options That Have Been Proposed

1)   Higher Fares:  The first alternative that many would suggest for raising additional funds for the transit system is to charge higher fares.  While certainly feasible in a mechanical sense, such an alternative would fail the effectiveness test.  The fares are already high.  Any increase in fares will lead to yet more transit users choosing to drive instead (for those for whom this is an option).  The increase in fare revenues collected will be less than in proportion to the increase in fare rates set.  And at some point, so many transit users will switch that total fare revenue would in fact decrease.

In the recently passed FY18 budget for WMATA, the forecast revenues to be collected from fares is $709 million.  This is down from an expected $792 million in FY17 despite a fare increase averaging 4%.  Transit users are leaving as fares have increased and service has deteriorated.  To increase the fares to try to raise an additional $650 million would require an increase of over 90% if no riders then leave.  But more riders would of course leave, and it is not clear if anything additional (much less an extra $650 million) would be raised. And this would of course be even more so if one tried to raise an extra $2.0 billion.

So as all recognize, it will not be possible to resolve the WMATA funding issues by means of higher fares.  Any increase in fares will instead lead to more riders leaving the system for their cars, leading to even greater road congestion.

2)  Increase the Sales Tax Rate:  Mayor Muriel Bowser of Washington has pushed for this alternative, and the recent COG Technical Panel concluded with the recommendation that  “the best revenue solution is an addition to the general sales tax in all localities in the WMATA Compact area in the National Capital Region” (page 4).  This alternative has drawn support from some others in the region as well, but is also opposed by some. There is as yet no consensus.

Sales taxes are already imposed across the region, and it would certainly be feasible to add an extra percentage point to what is now charged.  But each jurisdiction sets the tax in somewhat different ways, in terms of what is covered and at what rates, and it is not clear to what the additional 1% rate would be applied.  For example, Washington, DC, imposes a general rate of 5.75%, but nothing on food or medicines, while liquor and restaurants are charged a sales tax of 10% and hotels a rate of 14.5%.  Would the additional 1% rate apply only to the general rate of 5.75%, or would there also be a 1% point increase in what is charged on liquor, restaurants, and the others?  And would there still be a zero rate on food and medicines?  Virginia, in contrast, has a general sales tax rate (in Northern Virginia) of 6.0%, but it charges a rate on food of 2.5%.  Would the Virginia rate on food rise to 3.5%, or stay at 2.5%?  There is also a higher sales tax rate on restaurant meals in certain of the local jurisdictions in Virginia (such as a 10% rate in Arlington County) but not in others (just the base 6% rate in Fairfax County).  How would these be affected?  And similar to DC, there are also special rates on hotels and certain other categories.  Maryland also has its own set of rules, with a base rate of 6.0%, a rate of 9% on alcohol, and no sales tax on food.

Such specifics could presumably be worked out, but the distribution of the burden across individuals as well as the jurisdictions will depend on the specific choices made.  Would food be subject to the tax in Virginia but not in Maryland or DC, for example?  The COG Technical Panel must have made certain assumptions on this, but what they were was not explained in its report.

But it concluded that an additional 1% point on some base would generate $650 million in FY2019.  This is higher than the estimate made last October as part of the COG Panel work, where it estimated that a 1% point increase in the sales tax rate would raise $500 million annually.  It is not clear what the underlying reasons were for this difference, but the recent estimates might have been more thoroughly done.  Or there might have been differing assumptions on what would be included in the base to be taxed, such as food.

A 1% point rise in the sales tax imposed in the region would, under these estimates, then suffice to raise the minimum $650 million needed now.  But to raise $1.0 billion annually, rising to $2.0 billion a few years later, substantial further increases would soon be needed. The amount would of course depend on the extent to which local sales of taxable goods and services grew over time within the region.  Assuming that sales of items subject to the sales tax were to rise at a 3% annual rate in nominal terms (2% for inflation and 1% for real growth), and that one would need to raise $2.0 billion by 2030 (in terms of the prices of 2030), then the base sales tax rate would need to rise by about 2.2% points.  A 6% rate would need to rise to 8.2%.  A rate that high would likely generate concerns.

Thus while a sales tax increase would be effective in raising the amounts needed to fund WMATA in the immediate future, with a 1% rise in the tax rate sufficing, the sales tax rate would need to rise further to quite high levels for it to raise the amounts needed a few years later.  Whether such high rates would be politically possible is not clear.

Also likely to be a concern, as the COG Panel itself recognized in its report, is that the distribution of the increased tax burden across the local jurisdictions would differ substantially from what these jurisdictions contribute now to fund WMATA, as well as from what it estimates each jurisdiction would be called on to contribute (under the existing sharing rules) to cover the funding gap anticipated for FY17 – FY26:

Funding Shares:

FY17 Actual

FY17-26 Gap

From Sales Tax

DC

37.3%

35.8%

22.8%

Maryland

38.4%

33.5%

26.5%

Virginia

24.3%

30.7%

50.8%

Source:  COG Panel Final Report, pages 9 and 15.

If an extra 1% point were added to the sales tax across the region, 50.8% of the revenues thus generated would come from the Northern Virginian jurisdictions that participate in the WMATA compact.  This is substantially higher than the 24.3% share these jurisdictions contributed in WMATA funding in FY17, or the 30.7% share they would be called on to contribute to cover the anticipated FY17-26 gap (higher than in just FY17 primarily due to the opening of the second phase of the Silver Line).  The mirror image of this is that DC and Maryland would gain, with much lower shares paid in through the sales tax increase than what they are funding now.  Whether this would be politically acceptable remains to be seen.

Use of a higher sales tax to fund WMATA needs would also not lead to efficiency gains for the transportation system.  The sales tax on goods and services sold in the region would not have an impact on incentives, positive or negative, on decisions on whether to drive for your commute or to take transit.  It would be neutral in this regard, rather than beneficial.

Finally, and perhaps most importantly, sales taxes are regressive, costing the poor more as a share of their income than what they cost the well-off.  A sales tax rise would not meet the fairness test.  Even with exemptions granted for foods and medicines, poor households spend a high share of their incomes on items subject to sales taxes, while the well-off spend a lower share.  The well-off are able to devote a higher share of their incomes to items not subject to the general sale tax, such as luxury housing, or vacations elsewhere, or services not subject to sales taxes, or can devote a higher share of their incomes to savings.

Aside from the regressive nature of a sales tax, an increase in the sales tax to fund transit (and through this to reduce road congestion) will be paid by all in the region, including those who do not commute to work.  It would be paid, for example, also by retirees, by students, and by others who may not normally make use of transit or the road system to get to work during rush hour periods.  But they would pay similarly to others, and some may question the fairness of this.

An increase in the sales tax rate would thus be feasible.  And while a 1% point rise in the rate would be effective in raising the amounts needed in the immediate future, there is a question as to whether this approach would be effective in raising the amounts needed a few years later, given constraints (political and otherwise) on how high the sales tax rate could go.  The region would likely then face another crisis and dilemma as to how WMATA can then be adequately funded.  There are also political issues in the distribution of the sales tax burden across the jurisdictions of the region, with Northern Virginia paying a disproportionate share.  This would be even more of a concern when the tax rate would need to be increased further to cover rising WMATA funding needs.  There would also be no efficiency gains through the use of a sales tax.  Finally and importantly, a higher sales tax is regressive and not fair as it taxes a higher share of the income of the poor than of the well-off, as well as of groups who do not use transit or the roads during the rush hour periods of peak congestion.

3)  A Special Property Tax Rate on Properties Near Metro Stations

Some have argued for a special additional property tax to be imposed on properties that are located close to Metro stations.  The largest trade union at WMATA has advocated for this, for example, and the COG Technical Panel looked at this as one option it considered.

The logic is that the value of such properties has been enhanced by their location close to transit, and that therefore the owners of these more valuable properties should pay a higher property tax rate on them.  But while superficially this might look logical, in fact it is not, as we will discuss below.  There are several issues, both practical and in terms of what would be good policy.  I will start with the practical issues.

The special, higher, tax rate would be imposed on properties located “close” to Metro stations, but there is the immediate question of how one defines “close”.  Most commonly, it appears that the proponents would set the higher tax on all properties, residential as well as commercial, that are within a half-mile of a station.  That would mean, of course, that a property near the dividing line would see a sharply higher property tax rate than its neighbor across the street that lies on the other side of the line.

And the difference would be substantial.  The COG Technical Panel estimated that the additional tax rate would need to be 0.43% of the assessed value of all properties within a half mile of the DC area Metro stations to raise the same $650 million that an extra 1% on the sales tax rate would generate.  It was not clear from the COG Panel Report, however, whether the higher tax of 0.43% was determined based on the value of all properties within a half-mile of Metro stations, or only on the base of all such properties which currently pay property tax.  Governmental entities (including international organizations such as the World Bank and IMF) and non-profits (such as hospitals and universities) do not pay this tax (as was discussed above), and such properties account for a substantial share of properties located close to Metro stations in the Washington region.  If the 0.43% rate was estimated based on the value of all such properties, but if (just for the sake of illustration; I do not know what the share actually is) properties not subject to tax make up half of such properties, then the additional tax rate on taxable properties that would be needed to generate the $650 million would be twice as high, or 0.86%.

But even at just the 0.43% rate, the increase in taxes on such properties would be large. For Washington, DC, it would amount to an increase of 50% on the current general residential property tax rate of 0.85%, an increase of 26% on the 1.65% rate for commercial properties valued at less than $3 million, and an increase of 23% on the 1.85% rate for commercial properties valued at more than $3 million.  Property tax rates vary by jurisdiction across the region, but this provides some sense of the magnitudes involved.

The higher tax rate paid would also be the same for properties sitting right on top of the Metro stations and those a half mile away.  But the locational value is highest for those properties that are right at the Metro stations, and then tapers down with distance. One should in principle reflect this in such a tax, but in practice it would be difficult to do. What would the rate of tapering be?  And would one apply the distance based on the direct geographic distance to the Metro station (i.e. “as the crow flies”), or based on the path that one would need to take to walk to the Metro station, which could be significantly different?

Thus while it would be feasible to implement the higher property tax as a fixed amount on all properties within a half-mile (at least on those properties which are not exempt from property tax), the half-mile mark is arbitrary and does not in fact reflect the locational advantages properly.

The rate would also have to be substantially higher if the goal is to ensure WMATA is funded adequately by the new revenue source beyond just the next few years.  Assuming, as was done above for the other options, that property values rise at a 3% rate over time going forward (due both to growth and to price inflation), the 0.43% special tax rate would raise $900 million by 2030.  If one needed, however, $2 billion by that year for WMATA funding needs, the rate would need to rise to 0.96%.  This would mean that residential properties within a half mile would be paying more than double the property tax paid by neighbors just beyond the half-mile mark (assuming basic property tax rates are similar in the future to what they are now, and based on the current DC rates), while commercial rates would be over 50% more.  The effectiveness in raising the amounts required is therefore not clear, given the political constraints on how high one could set such a special tax.

But the major drawback would be the impact on efficiency.  With the severe congestion on Washington region roads, one should want to encourage, not discourage, concentrated development near Metro stations.  Indeed, that is a core rationale for investing so much in building and sustaining the Metro system.  To the extent a higher property tax discourages such development, the impact of such a special property tax on real estate near Metro stations would be to discourage precisely what the Metro system was built to encourage.  This is perverse.  One could indeed make the case that properties located close to Metro stations should pay a lower property tax rather than a higher one.  I would not, as it would be complex to implement and difficult to explain.  But technically it would have merit.

Finally, a special additional tax on the current owners of the properties near Metro stations would not meet the fairness test as the current owners, with very few if any exceptions, were not the owners of that land when the Metro system locations were first announced a half century ago.  The owners of the land at that time, in the 1960s, would have enjoyed an increase in the value of their land due to the then newly announced locations of the Metro stations.  And even if the higher values did not immediately materialize when the locations of the new Metro system stations were announced, those higher values certainly would have materialized in the subsequent many decades, as ownership turned over and the properties were sold and resold.  One can be sure the prices they sold for reflected the choice locations.

But those who purchased that land or properties then or subsequently would not have enjoyed the windfall the original owners had.  The current owners would have paid the higher prices following from the locational advantages near the Metro stations, and they are the ones who own those properties now.  While they certainly can charge higher rents for space in properties close to the Metro stations, the prices they paid for the properties themselves would have reflected the fact they could charge such higher rents.  They did not and do not enjoy a windfall from this locational advantage.  Rather, the original owners did, and they have already pocketed those profits and left.

Note that while a special tax imposed now on properties close to Metro stations cannot be justified, this does not mean that such a tax would not have been justified at an earlier stage.  That is, one could justify that or a similar tax that focused on the initial windfall gain on land or properties that would be close to a newly announced Metro line.  When new such rail lines are being built (in the Washington region or elsewhere), part of the cost could be covered by a special tax (time-limited, or perhaps structured as a share of the windfall gain at the first subsequent arms-length sale of the property) that would capture a share of the windfall from the newly announced locations of the stations.

An example of this being done is the special tax assessments on properties close to where the Silver Line stations are being built.  The Silver Line is a new line for the Washington region Metro system, where the first phase opened recently and the second phase is under construction.  A special property tax assessment district was established, with a higher tax rate and with the funds generated used to help construct the line.  One should also consider such a special tax for properties close to the stations on the proposed Purple Line (not part of the WMATA system, but connected to it), should that light rail line be built. The real estate developers with properties along that line have been strong proponents of building that line.  This is understandable; they would enjoy major windfall gains on their properties if the line is built.  But while the windfall gains could easily be in the hundreds of millions of dollars, there has been no discussion of their covering a portion of the cost, which will sum to $5.6 billion in payments to the private contractor to build and then operate the line for 30 years.  Under current plans, the general taxpayer would be obliged to pay this full amount, with only a small share of this (less than one-fourth) recovered in forecast fares.

While setting a special (but temporary) tax for properties close to stations can be justified for new lines, such as the Silver Line or the Purple Line, the issues are quite different for the existing Metro lines.  Such a special, additional, tax on properties close to the Metro stations is not warranted, would be unfair to the current owners, and could indeed have the perverse outcome of discouraging concentrated development near the Metro stations when one should want to do precisely the reverse.

4)  Other Funding Options

There can, of course, be other approaches to raising the funds that WMATA needs.  But there are issues with each, they in general have few advocates, and most agree that one of the options discussed above would be preferable.

The COG Technical Panel reviewed several, but rejected them in favor of its preference for a higher sales tax rate.  For example, the COG Panel estimated that it would be possible to raise their target for WMATA funding of $650 million if all local jurisdictions raised their property tax rates by 0.08% of the assessed values on all properties located in the region. But general property taxes are used as the primary means local jurisdictions raise the funds they need for their local government operations, and it would be best to keep this separate from WMATA funding.  The COG Panel also considered the possibility of creating a new Value-Added Tax (or VAT), a tax that is common elsewhere in the world but has never been instituted in the US.  It is commonly described as similar to a sales tax, but is imposed only on the extra value created at each stage in the production and sale process. But it would be complicated to develop and implement any new tax such as this, and it also has never been imposed (as far as I am aware) on a regional rather than national basis.  A regional VAT might be especially complicated.  The COG Panel also noted the possibility of a “commuter tax”.  Such a tax would have income taxes being imposed on a worker based on where they work rather than where they live.  But since there would be an offset for any such taxes against what the worker would otherwise pay where they are resident, the overall revenues generated at the level of the region as a whole would be essentially nothing.  It would be a wash.  There is also the issue that Congress has by law prohibited Washington, DC, from imposing any such commuter tax.

The COG Panel also looked at the imposition of an additional tax on motor vehicle fuels (gasoline and diesel) sold in the region.  This would in principle be more attractive as a means for funding transit, as it would affect the cost of commuting by car (by raising the cost of fuel) and thus might encourage, at the margin, more to take transit and thus reduce congestion.  Fuel taxes in the US are also extremely low compared to the levels charged in most other developed countries around the world.  And federal fuel taxes have not been changed since 1993, with a consequent fall in real, inflation-adjusted, terms. There is a strong case that the rates should be raised, as has been discussed in an earlier post on this blog.  But such fuel taxes have been earmarked primarily for road construction and maintenance (the Highway Trust Fund at the federal level), and any such funds are desperately needed there.  It would be best to keep such fuel taxes earmarked for that purpose, and separated from the funding needed to support WMATA.

E.  Summary and Conclusion

All agree that there is a need to create a dedicated source of funds to provide additional funds to WMATA.  While there are a number of issues with WMATA, including management and governance issues, no one disagrees that a necessary element in any solution is increased funding.  WMATA has underinvested for decades, with the result that the current system cannot operate reliably or safely.

Estimates for the additional funding required by WMATA vary, but most agree that a minimum of an additional $650 million per annum is required now simply to bring the assets up to a minimum level of reliability and safety.  But estimates of what will in fact be needed once the current most urgent rehabilitation investments are made are substantially higher.  It is likely that the system will need on the order of $2 billion a year more than what would follow under current funding formulae by the end of the next decade, if the system’s capacity is to grow by what will be necessary to support the region’s growth.

A mandatory fee on parking spaces for all commuters in the region would work best to provide such funds.  It would be feasible as it can be implemented largely through the existing property tax system.  It would be effective in raising the amounts needed, as a fee equivalent to $1.30 per day would raise $650 million per year under current conditions, and a fee of $3.50 per day would raise $2 billion per year in the year 2030.  These rates are modest or even low compared to what it costs now to drive.

A mandatory fee on parking spaces would also contribute to a more efficient use of the transportation assets in the region not only by helping to ensure the Metro system can function safely and reliably, but by also encouraging at least some who now drive instead to take transit and hence reduce road congestion.  Finally, such a fee would be fair as it is those of higher income who most commonly drive (in part because driving is expensive), while it is the poor who are most likely to take transit.

An increase in the sales tax rate in the region would not have these advantages.  While an increase in the rate by 1% point was estimated by the COG Panel to generate $650 million a year under current conditions, the rate would need to increase by substantially more to generate the funds that will be needed to support WMATA in the future.  This could be politically difficult.  The revenues generated would also come disproportionately from Northern Virginia, which itself will create political difficulties.  It would also not lead to greater efficiencies in transport use, other than by keeping WMATA operational (as all the options would do).  Most importantly, a sales tax is regressive (even when foods and medicine are not taxed), with the poor bearing a disproportionate share of the costs.

A special property tax on all properties located a half mile (or whatever specified distance) of existing Metro stations could also be imposed, although readily so only on such properties that are currently subject to property tax.  But there would be arbitrariness with such a rigidly specified distance being imposed, with a sharp fall in the tax rate for properties just across that artificial border line.  There is also a question as to whether it would be politically feasible to set the rates to such high rates as would be necessary as to address the WMATA funding needs of beyond just the next few years.

But most important, such a special tax on the current owners would not be a tax on those who gained a windfall when the locations of the Metro stations were announced many decades ago.  Those original owners have already pocketed their windfall gains and have left.  The current owners paid a high price for that land or the developments on them, and are not themselves enjoying a special windfall.  And indeed, a new special property tax on developments near the Metro stations would have the effect of discouraging any such new investment.  But that is the precise opposite of what we should want.  The policy aim has long been to encourage, not discourage, concentrated development around the Metro stations.

This does not mean that some such special tax, if time-constrained, would not be a good choice when a new Metro line (or rail line such as the proposed Purple Line) is to be built. The owners of land near the planned future Metro stops would enjoy a windfall gain, and a special tax on that is warranted.  Such a special tax district has been set for the new Silver Line, and would be warranted also if the Purple Line is to be built.  Those who own that land will of course object, as they wish to keep their windfall in full.

To conclude, no one denies that any new tax or fee will be controversial and politically difficult.  But the Metro system is critical to the Washington region, and cannot be allowed to continue to deteriorate.  Increased funding (as well as other measures) will be necessary to fix this.  Among the possible options, the best approach is to set a mandatory fee that would be collected on all commuter parking spaces in the region.

Productivity: Do Low Real Wages Explain the Slowdown?

GDP per Worker, 1947Q1 to 2016Q2,rev

A.  Introduction, and the Record on Productivity Growth

There is nothing more important to long term economic growth than the growth in productivity.  And as shown in the chart above, productivity (measured here by real GDP in 2009 dollars per worker employed) is now over $115,000.  This is 2.6 times what it was in 1947 (when it was $44,400 per worker), and largely explains why living standards are higher now than then.  But productivity growth in recent decades has not matched what was achieved between 1947 and the mid-1960s, and there has been an especially sharp slowdown since late 2010.  The question is why?

Productivity is not the whole story; distribution also matters.  And as this blog has discussed before, while all income groups enjoyed similar improvements in their incomes between 1947 and 1980 (with those improvements also similar to the growth in productivity over that period), since then the fruits of economic growth have gone only to the higher income groups, while the real incomes of the bottom 90% have stagnated.  The importance of this will be discussed further below.  But for the moment, we will concentrate on overall productivity, and what has happened to it especially in recent years.

As noted, the overall growth in productivity since 1947 has been huge.  The chart above is calculated from data reported by the BEA (for GDP) and the BLS (for employment).  It is productivity at its most basic:  Output per person employed.  Note that there are other, more elaborate, measures of productivity one might often see, which seek to control, for example, for the level of capital or for the education structure of the labor force.  But for this post, we will focus simply on output per person employed.

(Technical Note on the Data: The most reliable data on employment comes from the CES survey of employers of the BLS, but this survey excludes farm employment.  However, this exclusion is small and will not have a significant impact on the growth rates.  Total employment in agriculture, forestry, fishing, and hunting, which is broader than farm employment only, accounts for only 1.4% of total employment, and this sector is 1.2% of GDP.)

While the overall rise in productivity since 1947 has been huge, the pace of productivity growth was not always the same.  There have been year-to-year fluctuations, not surprisingly, but these even out over time and are not significant. There are also somewhat longer term fluctuations tied to the business cycle, and these can be significant on time scales of a decade or so.  Productivity growth slows in the later phases of a business expansion, and may well fall as an economic downturn starts to develop.  But once well into a downturn, with businesses laying off workers rapidly (with the least productive workers the most likely to be laid off first), one will often see productivity (of those still employed) rise.  And it will then rise further in the early stages of an expansion as output grows while new hiring lags.

Setting aside these shorter-term patterns, one can break down productivity growth over the close to 70 year period here into three major sub-periods.  Between the first quarter of 1947 and the first quarter of 1966, productivity rose at a 2.2% annual pace.  There was then a slowdown, for reasons that are not fully clear and which economists still debate, to just a 0.4% pace between the first quarter of 1966 and the first quarter of 1982.  The pace of productivity growth then rose again, to 1.4% a year between the first quarter of 1982 and the second quarter of 2016.  But this was well less than the 2.2% pace the US enjoyed before.

An important question is why did productivity growth slow from a 2.2% pace between the late 1940s and mid-1960s, to a 1.4% pace since 1982.  Such a slowdown, if sustained, might not appear like much, but the impact would in fact be significant.  Over a 50 year period, for example, real output per worker would be 50% higher with growth at a 2.2% than it would be with growth at a 1.4% pace.

There is also an important question of whether productivity growth has slowed even further in recent years.  This might well still be a business cycle effect, as the economy has recovered from the 2008/09 downturn but only slowly (due to the fiscal drag from cuts in government spending).  The pace of productivity growth has been especially slow since late 2010, as is clear by blowing up the chart from above to focus on the period since 2000:

GDP per Worker, 2000Q1 to 2016Q2,rev

Productivity has increased at a rate of just 0.13% a year since late 2010.  This is slow, and a real problem if it continues.  I would hasten to add that the period here (5 1/2 years) is still too short to say with any certainty whether this will remain an issue.  There have been similar multi-year periods since 1947 when the pace of productivity growth appeared to slow, and then bounced back.  Indeed, as seen in the chart above, one would have found a similar pattern had one looked back in early 2009, with a slow pace of productivity growth observed from about 2005.

There has been a good deal of work done by excellent economists on why productivity growth has been what it was, and what it might be in the future.  But there is no consensus.  Robert J. Gordon of Northwestern University, considered by many to be the “dean in the field”, takes a pessimistic view on the prospects in his recently published magnum opus “The Rise and Fall of American Growth”.  Erik Brynjolfsson and Andrew McAfee of MIT, in contrast, argue for a more optimistic view in their recent work “The Second Machine Age” (although “optimistic” might not be the right word because of their concern for the implication of this for jobs).  They see productivity growth progressing rapidly, if not accelerating.

But such explanations are focused on possible productivity growth as dictated by what is possible technologically.  A separate factor, I would argue, is whether investment in fact takes place that makes use of the technology that is available.  And this may well be a dominant consideration when examining the change in productivity over the short and medium terms.  A technology is irrelevant if it is not incorporated into the actual production process.  And it is only incorporated into the production process via investment.

To understand productivity growth, and why it has fallen in recent decades and perhaps especially so in recent years, one must therefore also look at the investment taking place, and why it is what it is.  The rest of this blog post will do that.

B.  The Slowdown in the Pace of Investment

The first point to note is that net investment (i.e. after depreciation) has been falling in recent decades when expressed as a share of GDP, with this true for both private and public investment:

Domestic Fixed Investment, Total, Public, and Private, Net, percentage of GDP, 1951 to 2015, updated Aug 16, 2016

Total net investment has been on a clear downward trend since the mid-1960s.  Private net investment has been volatile, falling sharply with the onset of an economic downturn and then recovering.  But since the late 1970s its trend has also clearly been downward. Net private investment has been less than 3 1/2% of GDP in recent years, or less than half what it averaged between 1951 and 1980 (of over 7% of GDP).  And net public investment, while less volatile, has plummeted over time.  It averaged 3.1% of GDP between 1951 and 1968, but is only 0.5% of GDP now (as of 2015), or less than one-sixth of what it was before.

With falling net investment, the rates of growth of public and private capital stocks (fixed assets) have fallen (where 2014 is the most recent year for which the BEA has released such data):

Rate of Growth In Per Capita Net Stock of Private and Government Fixed Assets, edited, 1951 to 2014

Indeed, expressed in per capita terms, the stock of public capital is now falling.  The decrepit state of our highways, bridges, and other public infrastructure should not be a surprise.  And the stock of private capital fell each year between 2009 and 2011, with some recovery since but still at almost record low growth.

Even setting aside the recent low (or even negative) figures, the trend in the pace of growth for both public and private capital has declined since the mid-1960s.  Why might this be?

C.  Why Has Investment Slowed?

The answer is simple and clear for pubic capital.  Conservative politicians, in both the US Congress and in many states, have forced cuts in public investment over the years to the current low levels.  For whatever reasons, whether ideological or something else, conservative politicians have insisted on cutting or even blocking much of what the United States used to invest in publicly.

Yet public, like private, investment is important to productivity.  It is not only commuters trying to get to work who spend time in traffic jams from inadequate roads, and hence face work days of not 8 1/2 hours, but rather 10 or 11 or even 12 hours (with consequent adverse impacts on their productivity).  It affects also truck drivers and repairmen, who can accomplish less on their jobs due to time spent in jams.  Or, as a consequence of inadequate public investment in computer technology, a greater number of public sector workers are required than otherwise, in jobs ranging from issuing driver’s licenses to enrolling people in Medicare.  Inadequate public investment can hold back economic productivity in many ways.

The reasons behind the fall in private investment are less obvious, but more interesting. An obvious possible cause to check is whether private profitability has fallen.  If it has, then a reduction in private investment relative to output would not be a surprise.  But this has in fact not been the case:

Rate of Return on Produced Assets, 1951 to 2015, updated

The nominal rate of return on private investment has not only been high, but also surprisingly steady over the years.  Profits are defined here as the net operating surplus of all private entities, and is taken from the national account figures of the BEA.  They are then taken as a ratio to the stock of private produced assets (fixed assets plus inventories) as of the beginning of the year.  This rate of return has varied only between 8 and 13% over the period since at least 1951, and over the last several years has been around 11%.

Many might be surprised by both this high level of profitability and its lack of volatility.  I was.  But it should be noted that the measure of profitability here, net operating surplus, is a broad measure of all the returns to capital.  It includes not only corporate profitability, but also profits of unincorporated businesses, payments of interest (on borrowed capital), and payments of rents (as on buildings). That is, this is the return on all forms of private productive capital in the economy.

The real rates of return have been more volatile, and were especially low between 1974 and 1983, when inflation was high.  They are measured here by adjusting the nominal returns for inflation, using the GDP deflator as the measure for inflation.  But this real rate of return was a good 9.6% in 2015.  That is high for a real rate of return.  It was higher than that only for one year late in the Clinton administration, and for several years between the early 1950s and the mid-1960s.  But it was never higher than 11%.  The current real rate of return on private capital is far from low.

Why then has private investment slowed, in relation to output, if profitability is as high now as it has ever been since the 1950s?  One could conceive of several possible reasons. They include:

a)  Along the lines of what Robert Gordon has argued, perhaps the underlying pace of technological progress has slowed, and thus there is less of an incentive to undertake new investments (since the returns to replacing old capital with new capital will be less).  The rate of growth of capital then slows, and this keeps up profitability (as the capital becomes more scarce relative to output) even as the attractiveness of new investment diminishes.

b)  Conservatives might argue that the reduced pace of investment could be due to increased governmental regulations, which makes investment more difficult and raises its cost.  This might be difficult to reconcile with the rate of return on capital nonetheless remaining high, but in principle could be if one argues that the slower pace of new investment keeps up profitability as capital then becomes more scarce relative to output. But note that this argument would require that the increased burden of regulation began during the Reagan years in the early 1980s (when the share of private investment in GDP first started to slow – see the chart above), and built up steadily since then through both Republican and Democratic administrations.  It would not be something that started only recently under Obama.

c)  One could also argue that the reduced investment might be a consequence of “Baumol’s Cost Disease”.  This was discussed in earlier posts on this blog, both for overall government spending and for government investment in infrastructure specifically.  As discussed in those posts, Baumol’s Cost Disease explains why activities where productivity growth may be relatively more difficult to achieve than in other activities, will see their relative costs increase over time.  Construction is an example, where productivity growth has been historically more difficult to achieve than has been the case in manufacturing.  Thus the cost of investing, both public and private, relative to the cost of other items will increase over time.  This can then also be a possible explanation of slowing new investment, with that slower investment then keeping profitability up due to increasing scarcity of capital.

One problem with each of the possible explanations described above is that they all depend on capital investments becoming less attractive than before, either due to higher costs or due to reduced prospective return.  If such factors were indeed critical, one would need to take into account also the effect of taxes on investment returns.  And such taxes have been cut sharply over this same period.  As discussed in an earlier blog post, taxes on corporate profits, for example, are taxed now at an effective rate of less than 20%, based on what is actually paid after all the legal deductions and credits are included.  And this tax rate has fallen steadily over time.  The current 20% rate is less than half the effective rate that applied in the 1950s and 1960s, when the effective rate averaged almost 45%.  And the tax rate on long-term capital gains, as would apply to returns on capital to individuals, fell from a peak of just below 40% in the mid-1970s to just 15% following the Bush II tax cuts and to 20% since 2013.

Such sharp cuts in taxes on profits implies that the after-tax rate of return on assets has risen sharply (the before-tax rate of return, shown on the chart above, has been flat).  Yet despite this, private investment has fallen steadily since the early 1980s as a share of GDP.

Such explanations for the reason behind the fall in private investment since the early 1980s are therefore questionable.  However, the purpose of this blog post is not to debate this. Economists are good at coming up with models, possibly convoluted, which can explain things ex post.  Several could apply here.

Rather, I would suggest that there might be an alternative explanation for why private investment has been declining.  While consistent with basic economics, I have not seen it before.  This explanation focuses on the stagnant real wages seen since the early 1980s, and the impact this would have on whether or not to invest.

D.  The Impact of Low Real Wages

Real wages have stagnated in the US since the early 1980s, as has been discussed in earlier posts on this blog (see in particular this post).  The chart below, updated to the most recent figures available, compares the real median wage since 1979 (the earliest year available for this data series) to real GDP per worker employed:

Real GDP per Worker versus Real Median Wage, 1979Q1 to 2016Q2, rev

Real median wages have been flat overall:  Just 3% higher in 2016 than what they were 37 years before.  But real GDP per worker is almost 60% higher over this same period.  This has critically important implications for both private investment and for productivity growth. To sum up in one line the discussion that will follow below, there is less and less reason to invest in new, productivity enhancing, capital, if labor is available at a stagnant real wage that has changed little in 37 years.

Traditional economics, as commonly taught, would find it difficult to explain the observed stagnation in real wages while productivity has risen (even if at a slower pace than before). A core result taught in microeconomics is that in “perfectly competitive” markets, labor will be paid the value of its marginal product.  One would not then see a divergence such as that seen in this chart between growth in productivity and a lack of growth in the real wage.

(The more careful observers among the readers of this post might note that the productivity curve shown here is for average productivity, and not the marginal productivity of an extra worker.  This is true.  Marginal productivity for the economy as a whole cannot be easily observed, nor indeed even be well defined.  However, one should note that the average productivity curve, as shown here, is rising over time.  This can only happen if marginal productivity on new investments are above average productivity at any point in time.  For other reasons, the real average wage would not rise permanently above average productivity (there would be an “adding-up” problem otherwise), but the theory would still predict a rise in the real wage with the increase in observed productivity.)

There are, however, clear reasons why workers might not be paid the value of their marginal product in the real world.  As noted, the theory applies in markets that are assumed to be perfectly competitive, and there are many reasons why this is not the case in the world we live in.  Perfect competition assumes that both parties to the transaction (the workers and employers) have complete information on not only the opportunities available in the market and on the abilities of the individual worker, but also that there are no costs to switching to an alternative worker or employer.  If there is a job on the other side of the country that would pay the individual worker a bit more, then the theory assumes the worker will switch to it.  But there are, of course, significant costs to moving to the other side of the country.  Furthermore, there will be uncertainty on what the abilities of any individual worker will be, so employers will normally seek to keep the workers they already have to fill their needs (as they know what these workers can do), than take a risk on a largely unknown new worker who might be willing to work for a lower wage.

For these and other reasons, labor markets are not perfectly competitive, and one should not then be surprised to find workers are not being paid the value of their marginal product.  But there is also an important factor coming from the macroeconomy. Microeconomics assumes that all resources, including labor resources, are being fully employed.  But unemployment exists and is often substantial.  Additional workers can then be hired at the current wage, without a need for the firm to raise that wage.  And that will hold whether or not the productivity of those workers has risen.

In such an environment, when unemployment is substantial one should not be surprised to find a divergence between growth in productivity and growth in the real wage.  And while there have of course been sharp fluctuations arising from the business cycle in the rate of unemployment from year to year, the simple average in the rate since 1979 has been 6.4%.  This is well in excess of what is normally considered the full employment rate of unemployment (of 5% or less).  Macro policy (both fiscal and monetary) has not done a very good job in most of the years since 1979 in ensuring there is sufficient demand in the aggregate in the economy to allow all workers who want to be employed in fact to be employed.

In such an environment, of workers being available for hire at a stagnant real wage which over time diverges more and more from their productivity, consider the investment decision a private firm faces.  Suppose they see a market opportunity and can sell more. To produce more, they have two options.  They can hire more labor to work with their existing plant and equipment to produce more, or they can invest in new plant and equipment.  If they choose the latter, they can produce more with fewer workers than they would otherwise need at the new level of production.  There will be more output per unit of labor input, or put another way, productivity will rise if the latter option is chosen.

But in an economy where labor is available at a flat real wage that has not changed in decades, the best choice will often simply be to hire more labor.  The labor is cheap.  New investment has a cost, and if the cost of the alternative (hire more labor) is low enough, then it is more profitable for the firm simply to hire more labor.  Productivity in such a case will then not go up, and may indeed even go down.  But this could be the economically wise choice, if labor is cheap enough.

Viewed in this way, one can see that the interpretation of many conservatives on the relationship between productivity growth and the real wage has it backwards.  Real wages have not been stagnant because productivity growth has been slow.  Labor productivity since 1979 has grown by a cumulative 60%, while real median wages have been basically flat.

Rather, the causation may well be going the other way.  Stagnant and low real wages have led to less and less of an incentive for private firms to invest.  And such a cut-back is precisely what we saw in the chart above on private (as well as public) investment as a share of GDP.  With less investment, the pace of productivity growth has then slowed.

As a reflection of this confusion, conservatives have denounced any effort to raise wages, asserting that if this is done, jobs will be lost as firms choose instead to invest and automate.  They assert that raising the minimum wage, which is currently lower in real terms than what it was when Harry Truman was president, would lead to minimum wage workers losing their jobs.  As a former CEO of McDonalds put it in a widely cited news report from last May, a $15 minimum wage would lead to “a job loss like you can’t believe.”   Fast food outlets like McDonalds would then find it better to invest in robotic arms to bag the french fries, he said, rather than hire workers to do this.

This is true.  The confusion comes from the widespread presumption that this is necessarily bad.  Outlets like McDonalds would then require fewer workers, but they would still need workers (including to operate the robotic arms), and those workers would be more productive.  They could be paid more, and would be if the minimum wage is raised.

The error in the argument comes from the presumption that the workers being employed at the current minimum wage of $7.25 an hour do not and can not possess the skills needed to be employed in some other job.  There is no reason to believe this to be the case.  There was no problem with ensuring workers could be fully employed at a minimum wage which in real terms was higher in 1950, when Harry Truman was president, than what it is now.  And average worker productivity is 2.4 times higher now than what it was then.

Ensuring full employment in the economy as a whole is not a responsibility of private business.  Rather, it is a government responsibility.  Fiscal and monetary policy need to be managed so that labor markets are tight enough to ensure all workers who want a job can get a job, while not so tight at to lead to inflation.

Following the economic collapse at the end of the Bush administration in 2008, monetary policy did all it could to try to ensure sufficient aggregate demand in the economy (interest rates were held at or close to zero).  But monetary policy alone will not be enough when the economy collapsed as far as it did in 2008.  It needs to be complemented by supportive fiscal policy.  While there was the initial stimulus package of Obama which was critical to stabilizing the economy, it did not go far enough and was allowed to run out. And government spending from 2010 was then cut, acting as a drag which kept the pace of recovery slow.  The economy has only in the past year returned to close to full employment.  It is not a coincidence that real wages are finally starting to rise (as seen in the chart above).

E.  Conclusion

Productivity growth is key in any economy.  Over the long run, living standards can only improve if productivity does.  Hence there is reason to be concerned with the slower pace of productivity growth seen since the early 1980s, and especially in recent years.

Investment, both public and private, is what leads to productivity growth, but the pace of investment has slowed since the levels seen in the 1950s and 60s.  The cause of the decline in public investment is clear:  Conservative politicians have slowed or even blocked public investment.  The result is obvious in our public infrastructure:  It is overused, under-maintained, and often an embarrassment.

The cause of the slowdown in private investment is less obvious, but equally important. First, one cannot blame a decline in private investment on a fall in profitability:  Profitability is higher now than it has been in all but one year since the mid-1960s.

Rather, one needs to recognize that the incentive to invest in productivity enhancing tools will not be there (or not there to the same extent) if labor can be hired at a wage that has stagnated for decades, and which over time became lower and lower relative to existing productivity.  It then makes more sense for firms to hire more workers with their existing stock of capital and other equipment, rather than invest in new, productivity enhancing, capital.  And this is what we have observed:  Workers are being hired, but productivity is not growing.

An argument is often made that if firms did indeed invest in capital and equipment that would raise productivity, that workers would then lose their jobs.  This is actually true by definition:  If productivity is higher, then the firm needs fewer workers per unit of output than they would otherwise.  But whether more workers would be employed in the economy as a whole does not depend on the actions of any individual firm, but rather on whether fiscal and monetary policy is managed to ensure full employment.

That is, it is the investment decisions of private firms which determine whether productivity will grow or not.  It is the macro management decisions of government which determine whether workers will be fully employed or not.

To put this bluntly, and in simplistic “bumper sticker” type terms, one could say that private businesses are not job creators, but rather job destroyers.  And that is fine.  Higher productivity means that a firm needs fewer workers to produce what they make than would otherwise have been needed, and this is important for ensuring efficiency.  As a necessary complement to this, however, it is the actions of government, through its fiscal and monetary policies, which “creates” jobs by managing aggregate demand to ensure all workers who want to be employed, are employed.

More on the High Cost of the Purple Line: A Comparison to BRT on the Silver Spring to Bethesda Segment

Comparison of Purple Line to BRT Cost, Silver Spring to Bethesda

This is a quick post drawing on a report in today’s Washington Post on the implementation of bus rapid transit (BRT) in Montgomery County, Maryland.  The article notes that one of the early BRT routes planned in the county would run from Burtonsville to Silver Spring down US Highway 29, with an estimated capital cost of $200 million.

This would be a distance of 10.2 miles, so the cost would be $19.6 million per mile on average.  This BRT line is currently slated to stop in Silver Spring, but it would be straightforward to extend it along East-West Highway for a further 3.7 miles to Bethesda. Assuming the same average cost per mile, the capital cost of this addition would be $72 million.

The current plan is for the Purple Line light rail line to cover this same basic route, connecting Silver Spring to Bethesda.  As I have discussed in earlier blog posts, the Purple Line is incredibly expensive, even if one ignores (as the official cost estimates do) the environment costs of building and operating the line (including the value of parkland destroyed, which is implicitly being valued at zero, as well as the environmental costs from storm water run-off, habitat destruction, hazardous waste issues, higher greenhouse gas emissions, and more).  The current capital cost estimate, following the service and other cuts that Governor Hogan has imposed to bring down costs, is $2.25 billion.  This also does not include the costs that Montgomery County will cover directly for building the Bethesda station and well as the cost of a utilitarian path to be built adjacent to the train tracks.  The Purple Line would also cost more to operate per rider than the Montgomery County BRT routes are expected to cost, so there is no cost savings from lower operating costs.

The Purple Line would be 16.2 miles long in total.  Using just the $2.25 billion cost figure, this comes to $139 million per mile.  This is extremely high.  Indeed, the Columbia Pike streetcar line in Arlington County, which was recently cancelled due to its high cost, would have cost “only” $117 million per mile despite it being built through a high density urban corridor for most of its entire route.

The distance from Silver Spring to Bethesda on the Purple Line will be 4.4 miles if it is built. This is longer than the direct route by road since it will follow a more indirect path passing up and around the direct route.  Assuming the cost of this 4.4 miles is the same on average as for the rest of the Purple Line (it might be higher due to the need to build some major bridges, including over Rock Creek), the cost would come to $612 million.

The choice therefore is between spending $612 million to build this segment of the Purple Line from Silver Spring to Bethesda, or spending $72 million by extending the BRT.  The Purple Line cost is 8.5 times as much, and government could save $540 million ( = $612m – $72m) by terminating the Purple Line in Silver Spring and using BRT service instead.

As an earlier blog post argued, new thinking is necessary if we are to resolve the very real transportation issues we face in this region.  This is one more example of what could be done.  A half billion dollar savings is not small.

 

Concrete Measures to Address Real Income Stagnation of the Poor and Middle Classes

Piketty - Saez 1945 to 2013, June 2015, log scale

I.  Introduction

The distribution of the gains from economic growth has gotten horribly skewed since around 1980, as the graph above shows (using a log scale so that distances on the vertical scale are equal relative changes).  Average real incomes of the bottom 90% of households have fallen by 5% in real terms since 1980, while the real incomes of the top 0.01% have grown by 325%.  This is astounding.  Something has changed for the far worse in recent decades.  The real incomes of these groups grew at roughly similar rates in the post-World War II decades leading up to 1980, but then diverged sharply.

An earlier post on this blog looked at the proximate factors which took substantial growth in GDP per capita (which grew at about the same pace after 1980 as it had before) down to median wages that simply stagnated.  As discussed in that post, this was principally due to a shift in distribution from labor to capital, and a shift within labor from the lower paid to the higher paid.  (Demographic effects, principally the increased participation of women in the labor force, as well as increases in the prices of items such as medical care relative to the prices of other goods, were also both important during this period. However, both have now become neutral, and are not factors leading to the continuing stagnation in recent years of median wages.)

The purpose of this blog post is to look at concrete policy measures that can be taken to address the problem.  The issue is not slow growth:  As noted above, per capita growth in GDP since 1980 has been similar to what it was before.  The problem, rather, is the distribution of the gains from that growth, which has become terribly skewed.

Discussion of this is also timely, as a consensus appears to be forming among political leaders in both major parties that something needs to be done about the stagnant incomes of the poor and middle classes.  Several of the candidates seeking the Republican presidential nomination have said they wish to make this a primary issue of the 2016 campaign.  This is good.  If they are serious, then they would support measures such as those laid out below.  I doubt they will, as they were strenuously opposed in the past to many of them, and indeed championed the changes in policy from the Reagan period onwards which are, at a minimum, associated in time with the deterioration in distribution seen in the chart above.  But one can hope.

This is a long blog post, as it discusses a long list of measures that could be taken to address the predicament of the poor and middle classes.  Many (although not all) of these policies have been reviewed in previous posts on this blog.  Thus the discussion here will in such cases be kept individually brief, with the reader encouraged to follow the links to the earlier blog posts for more substantive discussion of the points being made.  And the reader with limited time may wish to scan through the section headings, and focus on those topics of most interest.

II.  A Policy Program

A.  Labor Market

We start with the labor market, as it is fundamental.

1)  Raise the minimum wage:  The minimum wage is now less in real terms than what it was in 1950, during the presidency of Harry Truman.  This is amazing.  Or perhaps what is amazing is the argument made by some that raising it would price workers out of their jobs.  Real GDP per capita is now 3.75 times what it was then, and labor productivity has grown to 3.5 times what it was then.  But the minimum wage is now less in real terms than what it was then.  The minimum wage in 1950 was not too high to price low paid workers out of the market, and labor productivity is three and a half times higher now.

Obama has called for the minimum wage to rise to $10.10 per hour from its current $7.25 per hour (with this then indexed to the rate of inflation).  This would bring the minimum wage back only to where it was in the 1960s, a half century ago.  There is no evidence that such a rise will hurt low wage workers, and it would still leave a full time worker (at 40 hours a week, and with no vacation time, so 52 weeks per year times 40 hours per week = 2,080 hours per year) at such a minimum wage (2,080 x $10.10 = $21,008 per year) earning well less than what is needed to bring a family of four up to the poverty line ($24,250 per year for a family of four in 2015).  This is the minimum that should be done. Indeed, it should be more.

2)  Ensure predictable work hours:  Many workers, particularly in sectors such as retail and food service (whether fast food or traditional restaurants), are increasingly being required to accept “flexible” work hours, where their employer tells them only a short time ahead what days to come in, at what time to report, and for how many hours.  They might be told a few days before, or only the day before, or sometimes only on the day of possible work, whether they will be needed and should report to work.  This is sometimes called “just-in-time” scheduling (a term taken from just-in-time inventory management) or “on-call” scheduling, and managers are rewarded for what is seen as “optimizing” their use of labor.

But it plays havoc with the life of many workers.  They cannot take a second job in the evening to help make ends meet, if they do not know whether their primary job may sometimes require that they come in on short notice to work an evening shift.  Students cannot enroll in college classes in order to finish a degree if they do not know whether they might be called in during class time.  Parents and especially single mothers can have great difficulty in arranging for child care when their work schedules are unpredictable.  And an unpredictable number of weekly work hours, of perhaps 30 hours one week and 20 hours the next, makes budgeting impossible.

Recent developments in computer and communications technology have made on-call scheduling possible and increasingly common.  Establishments such as Starbucks can receive at some central office real-time information on coffee sales (from the cash registers, via an Internet connection), and together with other factors (weather forecasts; whether a convention is in town) can run sophisticated software algorithms predicting how many workers will be needed, exactly when, and which ones should be called in.

The problems this creates for low income workers has only recently come to be recognized.  An important spark was a story in the New York Times on August 13, 2014, on the impact on a worker at Starbucks.  This led Starbucks the very next day to announce it would change its practices, but a review a month later by another news organization raised questions as to whether anything of significance had really changed.

Starbucks can possibly be shamed into improving it working conditions, as it sells high-end coffee to those with significant disposable income.  And if the problem were solely with Starbucks and a few similar firms, a shaming approach might possibly be effective. However, while Starbucks may have become the symbol, the practice has unfortunately become increasingly widespread.  Furthermore, there is competitive pressure across firms to adopt such practices.  If Starbucks stops the practice, it may face coffee retailers who continue to “optimize” their use of labor through such practices, who can then take away business by charging less.  There is thus pressure on firms to impose the lowest standards they can get away with.

There is no federal law against such labor practices.  There may be state laws which might constrain the practices in some way in certain jurisdictions, but they are not widespread. What is needed is a legislated solution which will apply generally, and should be undertaken at the federal level.  This has been done before.

Abuses of labor had become common during the “gilded age” of the late 1800s / early 1900s (when income distribution was, not coincidentally, as bad as it is now).  Progressive Era and later New Deal reforms addressed some of the more egregious issues.  Work place safety and child labor laws were enacted, a minimum wage was established, and a 40 hour work week was set as the standard, where a worker is required to be paid overtime at the rate of “time and a half” (150% of their regular wage) for any hours worked beyond 40 in a week.

The issue of unpredictable work hours could be addressed similarly.  All workers would have a normal set of hours, defined individually and a function of their particular job.  They would always be paid for this set of normal hours.  But it is recognized that firms may have an unexpected need for additional workers at certain times (for example to substitute for a worker who is sick).  Workers could take on such additional shifts beyond their regular hours if they so chose, but as a financial inducement for the firm not to rely on such job calls, the firm would be required to pay 150% of normal wages for such time (as for overtime work).

This would lead firms to assess better when they need workers and when not, and build this into the standard work schedules.  Firms have little incentive to do this now, since the cost of unpredictable schedules is shifted onto the workers.  Note also that with this system there will not be an incentive to further split up jobs, as all workers will be assigned an individualized normal schedule.  The firms will gain no flexibility by splitting a 30 hour a week job into one of 20 hours and one of 10 hours.  They will still need to determine when workers are expected to be needed, in order to assign hours accordingly.  And to the extent they do not do this carefully, they will be penalized as they will then need to pay 150% of the wage when a worker needs to be brought in to cover those extra hours.

Predictability in hours is extremely important to low wage workers.  Indeed, it is quite possibly more important than raising the minimum wage.  It will shift some of the costs of unpredictability back onto the firm, where it had been until modern computer and communications technology (and lax labor laws that did not foresee this) allowed the firms to shift onto workers the cost of unpredictability.  Firms did this to raise profits, and business profits did indeed then rise while wage earnings stagnated.  This is precisely the issue that needs to be corrected.

3)  Manage policies to return the economy rapidly back to full employment when there are economic downturns:  While it would be ideal always to keep the economy at or close to full employment, economic downturns happen, most recently in 2008 in the last year of the Bush administration.  But when they do, the priority should be to return the economy to full employment as rapidly as possible.  Everyone can of course agree on this. Differences arise, however, on how to do it.

Monetary policy should be used to the extent possible, but there is a limit to how much it can do since interest rates cannot go below zero.  And they have been at essentially zero since late 2008, following the Lehman Brothers collapse.  The Fed can also try to reduce long term rates (such as through quantitative easing), and while its policy here has had some beneficial effect, it is also limited.

Monetary policy in a downturn such as that just experienced must therefore be supported also by expansionary fiscal policy.  Additional government spending is the most effective way to expand demand, as it does this directly.  Tax cuts can also help, but are less effective since a significant share of the reduction in taxes may be partly saved by households (and likely will be largely saved by higher income households).  There will thus be less stimulative effect per dollar from tax cuts than from direct spending.

Unfortunately, government spending was cut each year from 2010 up to 2014. This was the first time in at least 40 years that government has cut spending in a downturn.  Hence the economic recovery has been the slowest of any over that period, and only picked up in 2014, when government spending was finally allowed to increase.

Republicans have argued that to return the economy to full employment, one should instead reduce regulations.  But this makes no sense.  Regulations in the middle of a downturn are not far different from what they were before the downturn began, so how can they be blamed for the unemployment, and how would reducing them lead to less unemployment?  And if some regulations are wrong, one should change them, whether the economy has high unemployment or low unemployment.  Regulations do not serve as a macro policy to reduce unemployment, but rather serve a micro purpose to ensure the economy functions efficiently.

A slack labor market will have a direct effect on income inequality.  When there is available labor that is unemployed, a laborer will have little leverage to negotiate for higher wages. And unemployment has been higher, on average, in the three decades following 1980 than in the three decades before 1980.  A simple regression analysis suggests that if the average rate of unemployment after 1980 had simply matched what it had been before 1980, then the real incomes of the bottom 20% of households would not only have grown at a rate similar to how fast they had grown before, but also at a rate similar to that of the top 20% in the period after 1980.  That is, there would have been continued growth in the real incomes of the bottom 20% after 1980, instead of stagnation, and no increase in inequality relative to the top 20%.  Keeping the economy at close to full employment is critical to the poor and middle classes.

B.  Fiscal Policy

Fiscal policy is therefore an important instrument to keep the economy at full employment, or to return it to full employment from a downturn.  What specifically should be done?

1)  End the Congressional budgetary “pay-as-you-go” rules when in or recovering from a recession:  The Budget Enforcement Act of 1990 required that, as a budgetary rule, any increase in mandatory government spending or reductions in taxes must be “paid for” over the next five years as well as the next ten years by offsetting spending cuts or tax increases, so as to be budget neutral over these periods.  While the rules have been modified over the years, were not in place for a period during the Bush II administration, and were not always abided by (such as for the large tax cuts at the start of the Bush administration), they have acted in recent years to limit the government spending that was needed to recover from the downturn.

Over the course of a full business cycle, covering the full period over when the economy is at or close to full employment and when it is not, a limit on the size of the government deficit is warranted.  But by setting such a limit to apply identically and continuously both in a downturn and when the economy is booming, one limits the government expenditures that may be needed for a rapid recovery.  The rule should be suspended during any period when the economy is in a recession or recovering from one, to be replaced by a rule that applies over the course of the business cycle as a whole and which focusses on the government debt to GDP ratio rather than simply the government deficit.

2)  Stop cuts to important safety net programs such as food stamps and unemployment insurance, especially in a downturn:   Safety net programs such as food stamps (now called SNAP) and unemployment insurance are critically important to the poor and middle classes, and especially so in a downturn. They provide limited support to households who lost their jobs or other sources of income in the recession, due to no fault of their own.  (The 2008 downturn was a consequence of the reckless management of banks and other financial institutions in the US, and the policy decision of the Bush administration officials not to make use of their regulatory powers to limit it.  This led to very high bank leverage, a build-up of the housing bubble, and then collapse when the housing bubble burst.)

As unemployment rose and incomes fell, especially of the poor and near poor, many more households became eligible for food stamps and, having lost jobs, for unemployment insurance.  Expenditures on these and other safety net programs expanded, as they are designed to do in a downturn.  But Congressional Republicans then forced through cuts in the key safety net programs (e.g. unemployment insurance, and food stamps) by limiting eligibility, and have sought much more extensive cuts.

If one wishes to help the poor and near poor, one does not cut programs which help them directly, and especially not when they most need them.  These programs are also extremely efficient.  The food stamp program (SNAP) spends only 5% of its budget on administrative costs.  Few charities are anywhere close to as efficient, but at least some prominent Republicans think otherwise.  Former Congresswoman Michele Bachmann, at one point the leading contender for the Republican presidential nomination in 2012, asserted that 70% of food stamp funding went to “bureaucrats”.  This was absurdly wrong.

Safety net programs in the US, while efficient for the money spent, are however highly limited in how much they do spend.  US income inequality or poverty rates are actually not worse than those seen in other high income OECD economies in terms of wages and other income paid to workers. That is, the US capitalist market economy does not itself produce more unequal outcomes than those of other high income OECD economies.  The US is near the average in this across all the OECD countries.  But once one takes into account government taxes and transfers, the US turns out to be the very worst, both in terms of inequality (as measured by the Gini coefficient) and in terms of poverty (by the share of the population considered poor).  The problem is not that the US tax and transfer programs are not especially progressive:  They are in fact more progressive than in most countries.  The problem is that they are simply too small to matter.

3)  Implement a public infrastructure investment program:  American public infrastructure is an embarrassment.  Compared to other high income countries, US roads, bridges, mass transit, and other public structures, are clearly inadequate in scale, are in terrible condition due to inadequate maintenance, and constitute what is in effect a growing debt that will need to be repaid in the future when they finally have to be rebuilt (normally at much greater cost than if they had been maintained properly).

Public investment has been cut back especially sharply in recent years, whether measured as a share of GDP or simply in real per capita terms, as a consequence of Congressional cuts in the budget.  But while the cuts have been especially sharp since 2010, the problem is long-standing.  Amazingly, US non-defense public investment in structures in 2013 was less, in real per capita terms, than what it was in 1960, even though real per capita GDP almost tripled over this period.

It should therefore be no surprise that roads and bridges are in poor condition (with some bridges that have even collapsed), mass transit systems are in poor repair, and all are grossly inadequate to what we need.  This impacts especially the middle classes, who have to sit in traffic jams daily just to get to work.  Toll roads or tolled lanes built by private concessionaires have become fashionable in recent years to build roads that government is no longer willing to pay for, but they can be expensive.  The tolled lanes opened recently in Northern Virginia in the I95/I495 corridor have tolls that, for the entire length, could conceivably reach $40 or more per trip (and double that per day).  Rich people can afford this, but most of those with middle class incomes cannot.

Adequate public infrastructure is needed to raise productivity and for the economy to grow.  And the recent severe downturn should have been a time for a special effort to expand such investment, putting to work unemployed construction and other workers, in firms that had capacity to produce more than they could sell due to low demand in the downturn.  Furthermore, the government could have borrowed long term funds during this period at historically low rates, and even at times at negative real rates.  It was insanity not to utilize the unemployed labor and underutilized firms, financed by funds at low or even negative real cost, to build and repair infrastructure that the economy and especially the poor and middle classes desperately need.

But this was not done.  It instead will need to be done over the coming years (and at much higher cost, due to the lack of maintenance), when the economy is hopefully at full employment, interest rates have returned to normal levels, and anything extra spent on public infrastructure will need to come out of less being available for other goods and services, including for private investment.

4)  Increase public support to higher education:  At one time, students could pay for the total costs of attending a state university, including room and board, through work at just the minimum wage during summers and part time during the academic terms.  This is no longer the case.  In part this is due to the fall in the minimum wage in real terms from where it had been in the 1960s.  But it was also a consequence of the rising cost of university education (a result of Baumol’s cost disease) coupled with sharp cutbacks in the share of these costs covered by state support to their colleges and universities, shifting more of the cost onto students and their families.

This reduction in state support to their colleges and universities needs to be reversed, or soon state schools will in effect no longer exist.  They will have become essentially private schools catering to those who can afford them.  And while federal programs exist to help students (most importantly the Pell Grant program, which provides grants of up to $5,775 per year, in academic year 2015/16), they are limited and family income based.  The maximum Pell Grant goes only to the poorest households.

There are different ways to assist students from poor and middle class households to continue their schooling beyond high school.  President Obama, in this year’s State of the Union address, proposed for example that federal support be provided so that community colleges would stop charging tuition for students in good standing.  One would in essence be extending the availability of public schooling from 12 years now to 14 years.  The budget cost would be relatively modest at just $6 billion per year.  Others have suggested, constructively, that using such funds to expand the Pell Grant program would achieve the same aim, while assisting also those low income and middle class students who would do better by enrolling in a four-year college.

The response of the Republicans in Congress has, however, been in the opposite direction.  While rejecting Obama’s proposal for community colleges, the recent proposal from the House Budget Committee would instead freeze the maximum Pell Grant at $5,775 for at least the next ten years, thus leading to its erosion in real terms as prices rise (much as the minimum wage has eroded over time due to inflation).

C.  Tax Policy

Tax policy has a direct impact on income distribution.  But while most still accept the principle that taxes should be progressive (the principle that the rich should pay taxes at a higher rate than the poor), the tax system the US in fact has is regressive in many of its aspects.

1)  Stop the preferential tax treatment of income from wealth:  The wealthy pay a lower rate of taxes on their income from wealth than most of the population pays on their income from labor.  In terms of policy options to address inequality, little would be more straightforward than to end the practice of taxing income from wealth at lower rates than income from work.  Tax rates on all income groups could then be reduced, with the same total tax revenues collected.

The long-term capital gains tax rate on most assets is only 15% for most earners (there is an additional 3.8% for those households earning more than $250,000 bringing the rate to 18.8%, with this rising by a further 5% points to 23.8% for those earning $464,850 or more in 2015).  These rates on income from wealth are well below what is paid by most on income from labor (wages).  While the regular income tax rates (on wages) vary formally from 10% in the lowest bracket to 39.6% in the highest, one should add to these the social insurance taxes due.  These are often referred to loosely as Social Security taxes, but they include both Social Security at a 12.4% rate (up to a ceiling in 2015 of $118,500), and Medicare taxes at a 2.9% rate (with no ceiling).  Note also that while formally the employee pays half of this and the employer pays half, analysts agree that all of the tax really comes out of wages.

Once one includes social insurance taxes on wages, the effective tax rates on income from labor goes from 25.3% for the lowest bracket to 40.3% on those earning between $74,900 and $118,500, after which it drops down to 27.9% as the Social Security ceiling has been hit, and then starts to rise again.  The long-term capital gains rate is always below this even for the very richest, and normally far below.

To add further complication, preferential rates of 25% apply to long-term capital gains on certain commercial building assets (“Unrecaptured Section 1250” gains), and 28% apply to collectibles (such as fine art or gold coins) and certain small business stock. These are still well below what most pay in taxes on income from work.

This system not only worsens income inequality, but also creates complications and introduces distortions.  Many of those with high income are able to shift the categorization of their incomes from what for others would be wage income, to income which is treated as long-term capital gains.  Properly structured stock options, for example, allow CEOs and other senior managers to shift income from what would otherwise be taxed at ordinary income tax rates to the low rates for capital gains.  “Carried interest” does the same for fund managers.  With many fund managers earning over $100 million in a year, and indeed some earning over $1 billion, this preferential tax treatment of such extremely high earnings is perverse.

The reform would be simple.  All forms of income, whether from labor or from wealth, would be taxed at the same progressive rates that rise with total household income.  The one issue, which can be easily addressed, is that income from long-term capital gains should be adjusted for inflation, to put the gains all in terms of current year prices.  But this can easily be done by scaling up the cost basis based on the change in the general price level between when the asset was bought and when it was sold.  The IRS could supply a simple table for this.

2)  Reduce marginal effective tax rates on the poor:  Conservatives have long argued for cuts in marginal tax rates for the rich, arguing this would lead to faster growth (“supply-side economics”).  While there is no evidence that lower tax rates in recent decades have in fact led to faster growth, this was the stated rationale for the big tax cuts under Reagan and then Bush II.

Liberals have noted that if one were really concerned about high marginal tax rates, you would look at the high marginal effective tax rates being paid at the other end of the income scale – by the poor and those of moderate income.  Studies have found that these can range as high as 80 or even 100%.  The marginal effective tax rates take into account the impact of means-tested programs being phased out as one’s income rises.  When extra income is earned, you will pay not only a portion of this in taxes, but you will also lose a portion of benefits that are being provided (such as food stamps) in programs that phase out as income grows.  If the extra paid in taxes plus the amount lost in benefits matches additional earnings, the marginal effective tax rate will be 100%.

Conservatives have started to pay attention to this.  An opinion column in the Wall Street Journal last September by Senators Marco Rubio and Mike Lee (both Tea Party favorites) decried the 80 to 100% marginal effective tax rates that low income workers might face, arguing this can act as a strong disincentive to work.

In reality, the rates being faced by the poor are normally less, although still substantial. The issue is complex since the effective tax paid will depend not only on income, but on such factors as:  1)  Family composition (whether married and number and age of children); 2) Where one lives (what state and often what city or county); 3)  Particular benefit program qualification criteria (which will vary by program, and will often depend on many factors other than income); and 4) Whether the individual always enrolls in programs they are qualified for, as one may not be aware of certain benefit programs, or find the benefits to be too small to be worthwhile (because of difficulties and complications in enrolling, with these quite possibly deliberate difficulties in some jurisdictions).

The Congressional Budget Office, in a careful study issued in November 2012, concluded that the marginal effective tax rate for citizens with incomes of up to 450% of the federal poverty line averaged about 30% in 2012.  And under law as then in effect, this would rise to 32% in 2013 and 35% in 2014.  There was a good deal of variation within the average, with a marginal effective tax rate of as much as 95% (for example, for single mothers with one child with an income in the range of about $18,000 to $20,000, which was just above the poverty line for such a household).  But rates of 40% or more were not uncommon.

Such rates, even when not at the 80 to 100% extremes cited by Senators Rubio and Lee, are high.  Even at just 30% (the average) they are double what those who are far better off pay in long-term capital taxes.  While the conservative senators argue that such high marginal rates discourage work effort (there is in fact little evidence that this is the case – when you are poor, you are desperate for whatever you can get), such high marginal rates are in any case unfair.  If one wants to help those of low and moderate income, these effective tax rates should be reduced.

There are three ways, and only three ways, to reduce the marginal effective tax rates on the poor.  One is to get rid of the benefit programs for the poor altogether.  If they receive no food stamps to begin with, one has nothing to lose when incomes rise.  But this then penalizes the poor directly.

One could also phase out the programs more slowly, and pay for this by reducing the benefits to the poorest.  This is then a transfer from the poorest to the somewhat better off than the poorest.  Senators Rubio and Lee proposed this route (without explicitly calling it that) by reducing benefits such as food stamps and providing instead larger child tax credits for all households (including the rich).  This would be a transfer from the poorest to those of higher income (including the rich), and especially to those with large families.  The poorest would be penalized.

Finally, the third and only remaining option is to phase out the benefit programs more slowly.  This reduces the marginal effective tax rate that poor households will face if they are able to obtain jobs paying more than what they earned before.  It will not penalize the poorest, but it does of course require that budgets for the programs then rise.  But this will be support made available directly to the poor, and in particular to the working poor (as they are the ones facing the high marginal effective tax rates from additional earnings). If one is concerned about poverty and inequality, this is precisely the support that should be provided.

3)  Tax inherited wealth the same as any other wealth:  Wealth that is inherited enjoys significant tax benefits.  The only exception is wealth that is so large that it becomes subject to the estate tax (greater than $10.86 million for a married couple in 2015).  But few pay this due to that high ceiling, as well as since one can establish trusts and use other legal mechanisms to avoid the tax.  Indeed, in 2013 (the most recent year with available data), estate taxes were due on less than 0.2% of estates; the other 99.8% had no such taxes due.  This is down from over 6% of estates in the mid-1970s, due to repeated changes in tax law which narrowed and reduced further and further what would be due.

Any wealth that is passed along within the effective $10.86 million ceiling will never be subject to tax on capital gains made up to the point it was passed along.  That is, the cost basis is “stepped-up” to the value upon the date of death.  If one had purchased $100,000 of General Electric stock 40 years ago, the stepped-up value now would be roughly $3 million, and no taxes would ever be due on that $2.9 million of gain.

This is not fair.  Taxes on such wealth should be treated like taxes on any other wealth, and as argued in item #1 of this section above, all sources of income should be taxed the same (recognizing, importantly, that capital gains should be adjusted for inflation).

Some have argued that this cannot be done for inherited wealth since those inheriting the wealth may not know what the original cost basis was.  But this is not a valid excuse.  First of all, the two most important categories of wealth that are passed along through estates are homes and other real estate, and stocks and bonds and other such traded financial assets.  There are government land records on all real estate transactions, so it is easy (and indeed a matter of public record that anyone can look up) to find the purchase price of a home or other real estate.  And purchases of stocks, bonds, and other trade financial assets are done via a brokerage, which will have such records.

Second, if there are any other assets with special value (such as valuable coins or works of art) that will be passed along through an estate, the owner of those assets can include a record of their cost basis as part of the will or trust document that grants the asset to those inheriting the estate.

4)  Equal tax benefits for deductions should be provided for all income levels:  Under the current tax system, if a rich person in a 40% marginal income tax bracket makes a $100 contribution to some charity, then the government provides a gift to that rich person of $40 through the tax system, so that the net cost will only be $60.  If a middle class person in a 20% marginal tax bracket makes a $100 gift to the exact same charity, then the government will pay them back $20, for a net cost of $80.  And if a poor person in the 10% bracket makes a $100 gift to that same charity, the net cost to him will be $90.

This is enormously unfair.  There is no reason why deductions should be made more valuable to a rich person than to a poor person.  And there is an easy way to remedy it. Instead to treating deductions as subtractions from taxable income, one could take some percentage of deductions (say 20%) as a subtraction from taxes that would otherwise be due.  The $100 gift to the charity would then cost the rich person, the middle class person, and the poor person, the same $80 for each.

5)  Set tax rates progressively, and at the rates needed to ensure a prudent fiscal balance over the course of the business cycle:  I have argued above that all forms of income should be taxed similarly.  Whether you earn income from working or from wealth, two households with the same total income should pay taxes at the same rates.  This is not only for fairness.  It would also simplify the system dramatically.  And this is not only for the obvious reason that calculations are easier with one set of rates, but also because the current diverse rates interact between themselves in complex ways, which make computing taxes due a headache when there are different rates for different types of income.  Perhaps most importantly, it would eliminate the incentive to shift income from one category to another (e.g. from wage income to stock options) where lower taxes are due.  That creates distortions and wastes resources that should be used for productive activities.

There remains the question of what the new rates should be.  I do not have the data or the detailed tax models which would allow one to work that out, but a few points can be made.  One is that the new tax rates would be lower (at any given level of income) than what regular income tax rates (on wage income) are now.  This is because the tax rates on income from wealth (such as for capital gains, or inherited wealth) would be unified with the now higher tax rates on income from labor, so the new overall tax rates on regular income can be lower than before to yield the same level of tax revenues.  General income tax rates would come down.

Second, one should of course preserve the principle of progressivity in the tax code.  Rich people should pay taxes at a higher rate than poor people.  As Warren Buffitt argued eloquently in a New York Times column in 2011 (“Stop Coddling the Super-Rich”), there is no economic justification for the rich to pay taxes at lower rates than those with less income.  And it is grossly unfair as well.  Their secretaries should not pay at higher rates than the rich (fully legally) pay.

Third, one needs to recognize that the purpose of taxes is to raise revenues, and rates should thus be set at the levels required to cover government expenditures over time.  As argued above, the budgetary accounts should not be forced to balance each and every year, since fiscal policy is extremely important to move the economy back to full employment in an economic downturn (and fiscal policy is basically all that is available when interest rates are at the zero lower bound, as they have been since late 2008 in the US).

But over the full course of the business cycle, tax rates should be set so that the tax revenues collected suffice to keep the government debt to GDP ratio at a stable and reasonable level. This implies deficits when unemployment is high, and surpluses when the economy is close to full employment.  The economy was at full employment in 2006-07 (unsustainably so due to the housing bubble building up in those years) with the unemployment rate below 5% and as low as 4.4%.  Unfortunately, due to the Bush II tax cuts, the government’s fiscal deficit in those years was still significant.

Applying these principles, the tax rates needed for this would need to be worked out.  This can be done, but I do not myself have access to the data that would be required.  In the end, I suspect that regular income tax rates could be reduced substantially from what they are now, with this still sufficient to provide for adequate government expenditures over the course of the business cycle.  But the rich would pay more while those of low to moderate income would pay less, and whether this would then be possible politically is of course a different issue.

D.  Health Insurance

Access to health insurance is important.  A careful statistical analysis published in 2009 found that the likelihood of dying is higher for the uninsured than for the insured, and the lack of universal health insurance (as was then the case in the US) was leading to an estimated 45,000 more Americans dying each year than would be the case if they had health insurance.  This is greater than the number of Americans killed in action over the entire period of the Vietnam War.

The ObamaCare reforms have been a big step forward to making it possible for all Americans to obtain access to health care, but it remains under threat.  If Republicans are serious about helping the poor and middle classes, they should support the following.

1)  First, stop trying to block health care access for all:  ObamaCare, while not perfect (it reflects political compromises, and is based on the system of individual mandates first proposed by the conservative Heritage Foundation in 1989), is nonetheless working.  An estimated 16.4 million Americans now have health insurance coverage, which they did not have before ObamaCare became available.  And a Gallup poll found that there was a higher satisfaction rate among those obtaining health insurance policies through the ObamaCare exchanges, than among those with traditional health insurance policies (mostly via employers).

Despite this, Republicans continue to campaign aggressively to terminate ObamaCare, and have supported lawsuits in the courts to try to end this health care access.  A case now before the Supreme Court, with findings to be announced this month (June 2015), may declare that the federal government payments made through certain of the ObamaCare exchanges (those not run by the states) to those of low to moderate income, cannot be continued.  If the Supreme Court finds in favor of those opposed to such payments, these low and moderate income households will no longer be able to obtain affordable health insurance.

If the Republican presidential candidates are in fact in favor of helping the poor and middle classes, they should call for an end to the continued efforts by their Republican colleagues to terminate ObamaCare.  There are ways it could be improved (e.g. by taking more aggressive actions to bring down the cost of medical care to all in the US), but it is working as intended, and indeed better than even its advocates expected.

2)  Extend Medicaid in all states in the US:  The ObamaCare reforms built on the system of existing health insurance coverage.  Medicare would remain the same for those over age 65; employees in firms would normally pay for health insurance coverage through employer based plans; Medicaid would expand from covering (generally) those up to the poverty line to include also those with income up to 133% of the poverty line; and all those remaining without health insurance would be allowed to purchase coverage from private health insurers competing on internet-based health insurance exchanges, where those with incomes of up to 400% of the poverty line would receive federal subsidies to make such health insurance affordable.

The Medicaid expansion to cover all those with incomes of up to 133% of the poverty line was thus one of the building blocks in the ObamaCare reforms to enable all Americans to obtain access to affordable health care.  Because of its historical origins in health care programs for the poor that had traditionally been implemented in the states, Medicaid is implemented at the state level even though it is funded jointly with the federal government. For the Medicaid expansion to 133% of the poverty line, the federal government through the legislation setting up ObamaCare committed to funding 100% of the incremental costs in the first three years (2014 to 2016) with this then phased down to 90% in 2020 and thereafter.

Despite this generous federal funding, the Supreme Court decided in 2012, in its decision that also found ObamaCare in general to be constitutional, that Congress could not force the states to expand Medicaid to the higher income limits.  Thus the expansion became optional for the states, and as of April 2015, 21 states have decided not to.  The 21 states are mostly in the south or the mountain west, with Republican legislatures and/or Republican governors or mostly both.

Blocking this Medicaid expansion in these states is being done even though an expansion would cost little or nothing.  There is literally no state cost in the first three years as the federal government would cover 100% of the extra costs.  And while the federal share would fall then to a still high 90% for 2020 and thereafter, allowing more of the poor to be covered by Medicaid will reduce state costs.  The poor in this gap in coverage are forced to resort to expensive emergency room care, for which they cannot pay, when their health gets so bad that it cannot be ignored.  These costs are then partially compensated by the state.

A careful analysis for Virginia, undertaken for the state by Price Waterhouse Coopers, concluded that if Virginia opted in to the Medicaid expansion, the state would save $601 million in state budget costs over the period 2014 to 2022.  And this did not even include the indirect benefits to the budget from higher tax revenues, as a consequence of the additional jobs that would be created (nurses and other care providers, etc.) and the higher incomes of hospitals from less uncompensated care.

Denying affordable health care to the poor with incomes of between 100% and 133% of the poverty line is callous.  These are the working poor, and there really is no excuse.

3)  Ensure employers pay a proportionate share for health insurance for their part-time workers:  Employers have traditionally not allowed part time employees to obtain (through their wage compensation package) health insurance cover in employer-based plans.  There was a practical reason for this, as health insurance plans only provided cover in full.  It has been difficult to provide partial plans, so part time workers have traditionally gotten nothing.

This has now changed with the introduction of the ObamaCare health insurance market exchanges.  Those without health insurance coverage through their employers can now purchase a health insurance plan directly.  One could therefore now require that all employers make a proportionate contribution to the cost of the health insurance plans for their part time workers.  That is, for someone working half time (20 hours a week normally) the employer would contribute half of what that employer pays for the health insurance plans for their full time workers.

There would then be no incentive (and competitive advantage) for an employer to split a full time job with health insurance benefits into two half time jobs with no health insurance benefits for either worker.  Under the current system of normally no health insurance benefits for part time workers, employer are in essence shifting the cost of health insurance fully onto the worker and onto the federal government (and hence general taxpayer) if the worker earns so little that they are eligible for federal government subsidies to purchase health insurance on the exchanges.

The proportional employer contribution would be paid to the “account” of the worker in the same way (and at the same time) as Social Security and Medicare taxes are paid for the worker.  The funds from this account would then be used to cover part of the costs of the worker purchasing health insurance on the ObamaCare exchanges.  And if the worker was working in two (and sometimes three or more) part time jobs in order to make ends meet, the total paid in from all of the worker’s employers might well suffice to cover the cost of the health insurance plan in full.

4)  Allow competition from low cost public health insurance:  As part of the political compromises necessary to get ObamaCare passed in the face of steadfast Republican opposition, only private health insurers are allowed to participate in the ObamaCare health insurance exchanges.  Yet Medicare, which provides health insurance to all Americans age 65 and older, is far more efficient than private health insurance providers.  Giving Americans the option (not a requirement) to purchase Medicare administered health insurance on the ObamaCare exchanges would introduce much needed competition. There are often only a few insurers competing in the exchanges, which are all state-based (3 or fewer insurers in 15 of the 51 states plus Washington, DC, with only one insurer offering policies in West Virginia).

But even with multiple insurers competing on the state exchanges, the private insurers in general have high costs.  Private health insurers nationwide (and for all the policies they offer) have administrative costs plus profits equal to 14.0% of the health insurance benefits they pay out.  This is huge.  The same figure for Medicare is only 2.1%.  Medicare costs only one-seventh as much to administer as private health insurance.  And note these lower costs are not coming out of low payments to doctors and hospitals since the 14.0% and 2.1% are being measured relative to claims paid.

It would be straightforward to allow Medicare to compete on the health insurance exchanges.  One would require that Medicare charge rates sufficient (for the resulting client base, which will of course be younger than and generally healthier than Medicare’s senior citizens) to recover its full costs for such coverage.  But Medicare could use its existing administrative structure, including computer systems and contracts with doctors, hospitals, and other medical care providers at the rates that have already been negotiated.

Purchasing a Medicare administered health insurance policy on the exchanges would be fully optional.  No one participating in the exchanges would be required to buy it.  But with its lower costs, the competition Medicare would introduce into the exchanges could be highly effective in bringing down costs.

E.  Pensions

Probably the greatest failure of any social experiment of recent decades has been the switch of employer pension plans from the traditional defined benefit plans that were common up to the early 1980s, to defined contribution schemes (such as 401(k) plans) that have grown to dominance since the 1980s.  In doing this, employers not only shifted the investment, actuarial, and other risks on to the individual workers, but also typically reduced their matching funding shares significantly.

The end result is that workers are now typically woefully unprepared for retirement.  The funds accumulated in their 401(k) and IRA accounts for households (with head of household aged 55 to 64) with a 401(k) account, only amounted to $111,000 in 2013 (for the median household).  Such an accumulation, for households who will soon be in retirement, would suffice to provide only less than $400 a month by the traditional formulas, or $4,800 per year.

This is woefully inadequate, and these households will need to rely on what they can get from Social Security.  But Social Security payments are also not much.  For 2012 (the most recent year with such data), the median Social Security pension payment per beneficiary (age 65 or more) was just $16,799 per year, or per family unit (with head aged 65 or more) just $19,222 per year.  Furthermore, if nothing is done, the Social Security Trust Fund will run down to a zero balance in 2033 based on the most recent projections.

Fully funding Social Security is eminently solvable, as will be discussed below.  The annual cost (including for disability insurance) will rise from roughly 5% of GDP currently to 6% of GDP in 2030, as the baby boomers retire, or an increase of just 1% of GDP.  But what many do not realize is that on current projections, Social Security expenditures are then expected to remain stable (under current benefit rules) at that 6% of GDP for the foreseeable future, in projections that go out 75 years.  One can find 1% of GDP to save Social Security.  But the politicians in Washington will need to agree to do so.

There are several measures that need to be taken to ensure income security for the poor and middle classes in their old age:

1)  Raise, and certainly do not reduce, Social Security payments to those of low and modest income:  Social Security payments, while a crucial safety net, are low.  As noted above, the median payments in 2012 for those aged 65 and older were just $16,799 per beneficiary and $19,222 per recipient family (for the old age component). Such payments are minimal.  Yet these are just medians, meaning half of all recipients received less than even those figures.

Despite being so low, these Social Security payments are critical to many Americans. For the entire population of those aged 65 and older, Social Security accounted for half or more of their total regular income for two-thirds of the population (65% to be precise); Social Security accounted for 90% or more of their income for over a third of the population (36%); and Social Security accounted for 100% of their income for a quarter of the population (24%).  This is incredible, and shows the failure of the current pension systems in the US to provide a reasonable income in retirement for American workers.

The situation is, not surprisingly, worse for those of low to moderate income.  For the bottom 40% of the population by income, where 40% is by far not a small or insignificant share, Social Security accounted for half or more of their regular total income for 95% of them.  Social Security accounted for 90% or more of their income for three-quarters (74%) of them, and for 100% of their income for over half of them (53%).

Social Security benefits are by no means large.  Yet a large share of Americans depend on them.  They should not be cut, but rather should be raised, at least for those of low to moderate income who are critically dependent on them in their old age.

2)  Broaden the base for Social Security taxes to ensure the system remains fully funded:  As noted above, the Social Security Trust Fund will be depleted in 2033 (based on current projections, and including the disability component – while technically separate, the trust funds for old age and for disability insurance are normally treated together).  While some argue, with some justification, that the Trust Fund is fundamentally just an accounting tool, which can be “topped-up” with regular federal government transfers if necessary, there are also good reasons to stay with the Trust Fund rules.  They help keep the Social Security system out of routine Washington politics, and the temptation by conservatives to cut Social Security benefits in order to reduce the size of government.

Under the current Trust Fund rules, participation in the Social Security system is mandatory; taxes are paid on wage earnings (up to a ceiling of $118,500 in 2015); at a rate of 12.4% for Social Security pension and disability coverage for the employer and employee combined (excluding the 2.9% with no wage ceiling for Medicare); and upon retirement, these contributors will receive regular monthly payments from Social Security until they die, based on a formula which is linked to how much was paid into the Social Security Trust Fund during their working career (for the 35 years of highest inflation adjusted earnings).  The formula is complex, and “tilted” in the sense that those earning less will receive back somewhat more than they paid in, while those earning in the upper end of the taxable range will receive back somewhat less.  There is therefore some progressivity, but the degree of progressivity is limited.  Finally, the taxes paid into the fund will earn interest at the rate of the long-term Treasury bond yield of the time, so the taxes paid in to the fund grow over time with interest.

Based on these Trust Fund rules and on current projections of growth in worker earnings (and its distribution) and of what pay-outs will be as workers retire, the Trust Fund is expected to be depleted in 2033.  The basic cause is not that Social Security administration is inefficient and wasting funds.  It is, in fact, incredibly efficient, with a cost of administration of just 0.5% of benefits paid out (on the old age component).  Note this is not on assets, but on benefits paid.  As will be discussed further below, the typical expenses on savings in 401(k) and similar accounts will be 2 to 3% of assets each and every year.

The principal cause of the Trust Fund being depleted, rather, is that life expectancies have lengthened, and hence the period over which Social Security old-age payments are made have grown.  Crudely (and ignoring interest for this simple example), if life expectancy at age of retirement has grown by 50%, so that the number of years during which one will draw Social Security payments has grown by 50%, then the amount needed to be paid in to the Trust Fund to cover this lengthened life will need to grow by 50%.

Longer life expectancies are good.  The Social Security Trust Fund will run out of assets in 2033 not because the funds are being wasted (as noted above, administrative costs are incredibly low), but because people are living longer.  But this will need to be paid for.

But it is also critically important to recognize who is living longer.  As incomes have stagnated for those other than the rich since Reagan took office (see the chart at the top of this post), life expectancies for the poor and middle classes have in fact not increased substantially.  Rather, the overall life expectancy is rising principally because those at higher income levels are enjoying much longer life expectancies (in part precisely because their incomes have been growing so fast).

Specifically, life expectancy for a man at age 65 in the bottom half of earnings rose from 79.8 years in 1977 to 81.1 years in 2006, an increase of just 1.3 years.  But life expectancy at age 65 for a man in the top half of earnings rose from 80.5 years in 1977 (only 0.7 years longer than men in the lower half of the earnings distribution in that year) to 86.5 years in 2006, an increase of 6.0 years.  People are living longer, but it is mostly only those of higher income who are enjoying this.  Life expectancies have not changed much for those of low to moderate income.

It would therefore be perverse to penalize those of lower income to make up for the Trust Fund shortfall, when it is the lengthening of the life spans of those of higher income which is leading to the depletion of the Trust Fund.  Yet proposals to raise the retirement age, or to increase Social Security taxes on all, would charge the poor as much as the rich to make up for the Trust Fund deficit.

Rather than penalize the poor, Social Security taxes should be raised on those who are better off to pay for their increasing lifetime benefits.  And this is of course also precisely what one should want to do if one is concerned about inequality.

The obvious solution is to broaden the tax base for the Social Security tax to include incomes now excluded – which are also incomes that accrue to the rich.  First, note that the shortfall to be made up when the Social Security Trust Fund is expected to be depleted (if nothing is done) in the 2030s would require a 20% increase in revenues.  That is, following a transition between now and about 2030 as the baby boomer generation retires, the long term projection is that Social Security tax and other revenues will level off at about 5% of GDP, while Social Security obligations will level off at about 6% of GDP. That is, we will need revenues to increase by about 20% to make up the shortfall, to get to 6% of GDP from 5%.  (A more precise estimate, but from my blog post of a few years ago, is an increase of 19.4%).

Broadening the Social Security tax base would easily provide such an amount of funding. For wages alone, there is the current ceiling of $118,500 (in 2015) on wages subject to Social Security tax.  At this ceiling, only 83% of all wages paid are subject to this tax.  This is down from 89% in 1980 as an increasing share of wages are being paid to the very well off (those with wages above the ceiling).  If one were to tax 100% of wages rather than only the current 83%, one would in fact obtain the funds needed, as that alone would provide an increase of over 20% (100/83 = 1.205).

But broadening the base should not stop at simply ensuring the taxes paying for Social Security are paid by all wage earners equally, rich and poor.  There is no economic rationale why only wage income should be taxed for this purpose, and not also income from wealth.  And income from wealth is primarily earned by the wealthy.

Using figures from the National Income and Product (GDP) accounts, total private income (including not just wages, but also income from interest, dividends, rents, and so on, but excluding government transfers) in 2014 was 65% more than just wages alone.  (Note that while these income concepts from the GDP accounts are not the same as the income concepts defined in the income tax code, they suffice for the illustrative purposes here.)  A uniform tax rate of 7.5% rather than 12.4% (12.4/7.5 = 1.65), but on all forms of income and not just wage income, would thus suffice to generate the Social Security tax revenues to fund fully the Trust Fund for the foreseeable future.

But one important proviso should be noted.  Such tax rates (of either 12.4% on all wage income, or 7.5% on all forms of private income) would generate the revenues required to fully fund the system based on current benefit payment projections.  However, with benefit payments tied to one’s history of tax payments, one would also need to change the benefit payment formulae to reflect the broader tax base.  Otherwise the benefits due would also change to reflect the higher amounts paid in.

Finally, as noted above, current Social Security benefit payments are low and really need to be increased.  While I do not have the data and models that would be required to work out fully some specific proposal, the figures here can give us a sense of what is possible. For example, a possible balance might be to broaden the tax base to include all forms of income, but then to reduce the tax rate on this not all the way from 12.4% to 7.5%, but rather to the halfway point of 10%.  But this 10% rate would then suffice to permit a one-third increase in overall benefits (10/7.5 = 1.333), which one should want also to concentrate on those of low to moderate income.  The overall tax rate would be cut, but in broadening the tax base to all forms of income one could support a significant increase in benefits.  Overall taxes paid would be higher (everything has to add up, of course), and while low and moderate income earners would mostly see a reduction in the taxes they owe, richer individuals would pay more.

3)  Require that 401(k) plan administration fees are paid for by the entity choosing the provider:  Turning from Social Security to private pension plans, where defined contribution plans (401(k) plans and similar) are now the norm, the key issue is to keep fees low.  Unfortunately they are extremely high, and take away a large share of the investment returns on the funds saved.

One cause of this is that the fees being received by the financial advisors are often hidden in various ways, and charged in different ways by different entities.  There is no standard, and average levels are not made publicly available to provide a basis for comparison. The fees charged will also vary sharply depending on the type of investment product being used (i.e. direct equities, mutual funds of different types, insurance products including annuities, various complex products, and more).  Finally, while individual fees might sometimes appear to be low, they are imposed at several layers in the investment process, and in total come to very high levels.  Not surprisingly, according to one survey 93% of the respondents dramatically underestimated what their cumulative 401(k) fees will come to.

Estimates for what the average fees in fact now are vary.  At the base will be the fees charged by a plan administrator, chosen by the firm of the employee and responsible for the record keeping, for allocating the investments as chosen by the plan participant, and so on. These fees will vary depending on the size of the firm and the deal negotiated on behalf of the employee, but one set of estimates are that these fees average 0.5% of plan assets annually in large firms (plan sizes of $100 million or more), 0.9% in medium size firms (plan assets of $10 million to $100 million) and 1.4% in small firms (plan assets of less than $10 million).

On top of these base administrative fees one will then have to pay the fees charged by the investment vehicles themselves.  Mutual funds are perhaps most common, and the expense fees on these (normally subtracted from the investment returns, so the amount being paid will not be obvious) average about 1.1%.  But they can range widely, from less than 0.1% for standard index funds, to over 2%.  On top of this, one has the cost of buying and selling the investments (whether mutual funds or equities or the other products), which some estimate may average 1.0%.  But these too can vary widely.

Finally, one may have individual fees on top of these all, which depend on the services being requested (whether investment advice, or entry and exit fees, or loan advances against balances in the worker’s pension accounts, and more).  These will vary widely, but one estimate is that the median might be an annual cost of about 0.7% of assets.

The total amount lost in fees each year can therefore easily be in the range of 2 to 3% of assets, and 2.5% is a commonly taken average.  Note also that these fees (or at least most of them) will be taken by the financial managers regardless of whether the investments perform well in any given year.

Over time, these fees will take for the financial administrators a large share of the investment returns that were intended for worker retirement.  A simple spreadsheet calculation, for example, will show that over a 40 year time horizon, where (for simplicity) equal amounts are set aside each year for retirement, with an assumed 5% real rate of return but financial fees of 2.5% a year, the financial cost will have taken away by the 40th year 45% of what would otherwise be the balance saved.  (It will not equal 50% because of the way compounding works.)  This is 90 times the total fees that would be charged by Social Security for the old-age support it provides, which as noted above is only 0.5% of benefits paid out.  Note that I am not arguing that the private fees should be the same as what the cost is for Social Security to administer the accounts.  Social Security is an extremely efficient system.  The specific share taken by financial fees will also depend, of course, on what is assumed for the rate of return and other parameters. But anything like 90 times as much is a lot.

Furthermore, many of the fees being charged on the private pension accounts will continue into the retirement years, so the final amount paid out in benefits will be even less.  Assuming there will be 20 years of retirement following the 40 years of work, with returns and fees continuing as before, the annual amount that could be paid out to the worker in retirement will be only 44% of what would be paid out if the fees were at the 0.5% (of benefits only) of Social Security, a reduction of 56%!

Fees are therefore hugely important, but the worker can manage (through the decisions they make) only some of them.  In particular, the basic plan administration fees for managing the accounts are made at the firm level.  When firms had defined benefit pension plans, these costs were included in what the firm covered.  They kept track of, or contracted out to some specialist, the individual payment obligations and other bookkeeping.  But when firms switched to defined contribution schemes, most commonly 401(k)s, the firms chose to incorporate the costs of the plan administration into charges against the individual account balances, thus shifting these on to the workers.  But the workers had no choice between plan administrators:  They were chosen by the firm.  And with the firm not bearing the cost, the firm might not worry so much about what the cost was.  Rather, they might choose a plan administrator based on how good they were at making sales pitches, or whether they did other work for the firm where they might extend a discount, or based on who provided the fanciest annual conferences for their clients (the decision maker in the firm) at some resort in Hawaii.

Some firms of course make a sincere effort to choose the best balance between plan administrator cost and performance, but others do not.  But to resolve this, a simple reform would be to require that firms pay the cost of plan administration directly from the firm’s accounts, and not out of the pension savings of the workers.  This would return to what had normally been done with the traditional defined contribution pension plans, where firms had an incentive to ensure plan administration costs were kept low.

4)  Require low cost investment options be included in 401(k) and similar schemes:  One can also keep financial fees down by investing in low cost investment options, such as simple and standard index funds.  The Vanguard S&P 500 Index Fund, and now even the Vanguard Total Stock Market Index Fund, each charge an annual expense fee of just 0.05%.  This is far below the 1 to 2% charged by most managed mutual funds.

Such low cost options are included in many 401(k) plans, but unfortunately not all.  Such options should be required in all.

5)  Provide an option of investing in “Social Security Supplemental Accounts”:  One very low cost investment option would be to invest a portion of one’s 401(k), IRA, and similar balances, into a supplemental account managed by Social Security, which would then be paid out along with one’s regular Social Security checks.  Accounts would earn a return equal to the long-term US Treasury bond rate, the same as existing Social Security balances do now.  The option would take advantage of the extremely high efficiency of the Social Security Administration.

With the far lower cost of Social Security relative to what private financial institutions in the US charge, the net return on such investments would be attractive.  Given historical long term bond interest rates and the differences in fees, the returns would be comparable to what might be earned in pure equity investments, but far less volatile.  And the returns would be far better than what on average has been earned in actual 401(k) and IRA accounts, given how they have historically been managed.

The Social Security Supplemental Account would be purely voluntary.  No one would be forced to put a share of their 401(k) or IRA assets into them.  But this would be an attractive option for at least a share of the plan assets, earning a comparable return (after fees) but with far less risk.  It would be especially attractive to middle class families who may have significant but not huge assets in their retirement accounts, who are seeking to ensure a safe retirement.

F.  Conclusion

The returns to economic growth have become horribly skewed since Reagan took office, but there is much that could be done to address it.  This blog post has discussed a number of actions that could be taken, and if the Republican (and Democratic) candidates are truly concerned about the direction of income inequality, there is no shortage of measures to consider.  And this is not an all or nothing set of actions:  While complementary and mutually reinforcing, taking some actions is better than none, and more is better than some.  Nor are they in general administratively difficult to do.  Most are straightforward.

But action is clearly needed.

 

The Purple Line: New Thinking is Needed to Address Our Transportation Problems

The Washington Post in recent weeks has published a number of pieces, both articles and editorials, on purportedly massive economic benefits from building the Purple Line (a 16 mile light rail line in the Maryland suburbs of Washington, DC).  See, among others, the items from the Post herehere, here, and here.  Their conclusion is based on uncritical acceptance of the results of a consultant’s report which concluded that the funds invested would generate a return (in terms of higher incomes in the region) of more than 100% a year.  But transit projects such as this do not generate such huge returns.  This should have been a clear red flag to any reporter that something was amiss.

As discussed in an earlier post on this blog, the consultant’s report was terribly flawed. There were obvious blunders (such as triple counting the construction expenditures) as well as more complex ones.  While one should not expect a reporter necessarily to have the expertise to uncover such problems, one would expect a good reporter and news organization to have sought out the assessment of a neutral observer with the necessary expertise.  This was not done.

The consultant’s report was so badly done that one cannot conclude anything from it as to what the economic impact would be of building such a rail line.  But more importantly, the report did not even ask the right question.  It looked at building the Purple Line versus doing nothing.  But no one is advocating that we should do nothing.  We have real transportation issues, and they need to be addressed.

The proper question, then, is what is the best use of the scarce funds we have available for public transit.  If an alternative approach can provide better service for more riders at similar or lower cost, then the impact of building the Purple Line versus proceeding with that alternative is negative.  As discussed below and as an illustration of what could be done (there are other alternatives as well), instead of the Purple Line one could provide free bus service on the entire county wide local bus systems.  These bus systems cover poor communities as well as the rich (the Purple Line will pass through some of the richest zip codes in the nation), and for the cost of the Purple Line the bus systems could be expanded to provide free bus service to as many as four times the ridership the Purple Line is projected to carry (with such projections, based on past experience, likely to prove optimistic).  The local bus systems already carry twice the projected ridership of the Purple Line.

With the aim of providing an alternative view to the discussion being carried, I submitted the attached to the Post for consideration as a local opinions column.  The Post decided, however, not to publish it, so I am making it available here.

 

—————————————————————————————————-

Solving Our Transportation Problems Will Require New Thinking, Not Old White Elephants

As Governor Hogan comes to a decision on the Purple Line, there is increasing pressure from proponents arguing the 16 mile light rail line will yield huge developmental benefits.  A recently released and highly publicized study by the consulting firm TEMS concluded the line would raise incomes by $2.2 billion a year, for a capital cost that TEMS took to be just $1.9 billion (the cost estimate is now higher, at close to $2.5 billion).  Thus the return, TEMS says, will be well in excess of 100% a year.

As the saying goes, when something sounds too good to be true, it usually is.  The TEMS study was badly flawed.  More importantly, it did not address the right question.  What we should be asking is how best to address our very real transportation needs, given the limited public resources available.

There are numerous problems with the TEMS study.  To start with more obvious blunders:  It estimated impacts during the construction period as if the entire Purple Line would be built in Montgomery County, would be built again in Prince George’s, and built again in Washington, DC (even though it will not even touch Washington).  That is, it triple counted the construction expenditures and therefore its income and jobs impacts.  It also assumed that all the inputs (other than the rail cars) would be sourced in Washington or these counties.  But steel rail cannot come from here:  Neither Washington nor its suburbs have any steel mills.

There were more fundamental problems as well.  Among them was the fallacy of cause and effect.  In their statistical analysis, TEMS found that higher income neighborhoods are associated with lower transportation costs.  From this they jumped to the conclusion that lower transportation costs led to those higher incomes.  That is not the case.  Rich people live in Georgetown and, being close to downtown, transportation costs there are relatively low.  But moving to Georgetown does not suddenly make you rich due to low transportation costs.  Rather, one can afford to buy a home in Georgetown if you are already rich.

Perhaps the most basic problem is that TEMS assumed it was either the Purple Line or nothing.  But no one is advocating doing nothing.  We face real transportation issues, and they need to be addressed.  Unfortunately, alternatives have not been seriously examined.  Part of the problem has been narrow-minded thinking that has failed to consider broader alternatives than solely a line on this fixed corridor.

As an example of what might be done, consider the locally run RideOn and TheBus systems in these counties.  These two systems already carry double the projected ridership of the Purple Line.

The annual operating cost of the Purple Line is expected to be $55 million a year.  This is in addition to the $2.5 billion capital cost.  Taking just half of that annual operating cost, net of fares expected to be collected due to the Purple Line, one could cover the full amount currently collected in fares on the entire county-wide RideOn and TheBus systems.  That is, one could provide free bus service on the entire systems for just half of the cost of operating the Purple Line.

Some might say that with zero fares, there would be the “problem” that many more riders will want to take these buses.  But that would be fantastic.  One would need to cover the additional costs, but this could be done.  Note first that filling empty seats on buses costs nothing, and there are a lot of empty seats now.  There is then the second half of the annual operating cost of the Purple Line, and finally the $2.5 billion capital cost.  Taken together, these funds could cover the full costs (including costs covered in county and state budgets) of doubling the scale of the RideOn and TheBus systems.

We can therefore have the Purple Line, serving riders on a narrow 16 mile corridor that runs through some of the most affluent areas of these counties, or for the same cost provide free bus service on systems that could carry four times as many riders.  Furthermore, the bus systems serve not just affluent areas but also the poorest communities of the counties.  For the poor, earning at or close to the minimum wage with perhaps two jobs to get by, daily bus fares of $6 or $8 or more are not insignificant.

We need to be open to broader options for how to address our transportation crisis.  The debate on the Purple Line has not done that.  And by treating the issue as the Purple Line or nothing, proponents are increasing the likelihood that the outcome will be nothing.

—————————————————————————————————

Note on Sources:

a)  The TEMS consultant report, March 2015, commissioned by Montgomery County, Prince George’s County, and the Greater Washington Board of Trade.

b)  Current cost estimates for building and operating the Purple Line, along with ridership projections, are from the most recently issued Federal Transit Administration Purple Line Profile Sheet, November 2014.

c)  Cost and ridership data for the local transit systems (RideOn and TheBus) are from the National Transit Database of the Federal Transit Administration.