Tag Archives: entitlements

Creating an Attractive Retirement Savings Option, While Funding Social Security for a Further Generation

Posted on by

Social Security Outlays, Revenues, and Deficit, CBO, 1985-2086

A.  Introduction

This post will present a reform which would create an attractive, and wholly voluntary, new retirement savings option, which would also fund the US Social Security system for an additional generation.  And it would do this without cuts in benefits or increases in Social Security or other taxes.  While not a permanent fix for funding Social Security (where the issue is fundamentally a result of longer life expectancies but a largely unchanged work life), it could extend the lifetime of the Social Security Trust Fund without reducing benefits or increasing taxes by a further two decades or more.

Social Security is the most successful US social program of the last century.  The poverty rate of the elderly was extremely high before Social Security.  Because of Social Security it is now greatly reduced, to levels similar to that of other population groups.  While the benefits Social Security pays are not generous, and are far below the support provided to the elderly among the other advanced economies of Europe, Japan, and elsewhere, Social Security benefits have been sufficient to provide a minimum safety net of income that has been adequate to keep most elderly out of poverty.

But there is a need to ensure Social Security will remain sustainable.  As shown in the graph above, annual Social Security benefits paid exceed the inflow from Social Security taxes.  While there is a Trust Fund, and it is currently a sizable 17% of GDP, the Congressional Budget Office projects that if nothing is done, this Trust Fund will run down to zero by 2035.

At that time, and if one restricts the options to the current structure, either Social Security tax revenues would need to be scaled up (by about 23%, raising the tax rate on wages from the current 12.4% to about 15.2%, if nothing is done prior to that point to prepare for it), or benefits would have to be scaled back (not to zero, but by about 19% in the CBO forecast).  Raising the tax rate to 15.2% is not a disaster, and simply reflects the very desirable longer life expectancies we now enjoy.  And such a wage tax would still be far less than the rates paid in Europe.

One could also cover the roughly 1% of GDP deficit in the flows (from the 2030s to about 2060) from general tax revenues.  There is no requirement that all the support to the Social Security system has to come from wage taxes.  And this 1% of GDP is only about one-third to one-quarter or less (the loss rises over time) of the revenues that have been lost as a result of extending the Bush tax cuts for all but the top 2% of Americans.  If we can afford the Bush tax cuts being extended, we can afford one-third or one-quarter of it to save Social Security.

But any tax increases are anathema to Republicans.  In this blog post, I will review a different type of approach, which would create an attractive and wholly voluntary new retirement savings option.  As an additional benefit, it would provide funding for Social Security that would likely suffice to extend the life of the Social Security Trust Fund by a further two decades or more, without resorting to higher taxes or cuts in benefits.

B.  The Impact of Investment Fees

Social Security is an extremely efficient program.  Its administrative costs in the retirement program side (keeping the benefits for the disability program separate) are only 0.6% of the amount paid out annually in benefits (see 2012 Trustees Report, Table II.B.1.).  This is tiny.  Social Security is efficient for two primary reasons.  The most important is that it is structured as a simple program.  Revenues are collected through the tax system by a tax on wages, and these funds are then invested straight into (and only into) US Treasury bonds.  Second, it operates on an enormous scale, and such a scale while keeping costs low is possible precisely because of the simplicity of the design.

In contrast, retirement accounts that are run through private fund managers are highly fragmented, are structured with multiple investment options where choices must then be made on how much to invest in each, where expensive financial professionals demand high fees to administer these funds and to make recommendations (and sometimes decisions) on the investment choices, and where the high fragmentation inhibits scale economies.  Profits among the fund management companies can also be high.

As a consequence, the total fees on 401(k)’s, IRAs, and similar directed retirement investment programs (defined contribution schemes) can be extremely high.  The fees are assessed at several levels, including (for 401(k)’s) at the level of the individual plan for some firm, and at the level of the investment management company (whether through a mutual fund, through investment vehicles sold by life insurance companies, or by direct investments in the markets).  The fees add up, and can easily reach 2% of assets annually and beyond, depending on the size of the company one is working for, and the nature of the investments being made (whether equities, bonds, life insurance company products, money market funds, and so on).

These high fees have a dramatic impact on reducing the returns on the investments being made.  Taking the example of a pure equity investment, the total return on investment in the S&P500 Equity Index over the 50 years 1962-2012 was 9.7% annually (in nominal terms).  Adjusting for CPI inflation of 4.1% annually over this period, the real return was 5.4% (note that the 5.4% is not a simple subtraction of 4.1% from the 9.7% due to interaction effects:  take my word for this, or review your High School algebra).  But after a 2.0% annual fee on assets, the return would be reduced from 5.40% to only 3.31% for someone who starts saving for retirement at age 45 and lives to 92 (or 3.30% for someone who starts saving at age 22).  In contrast, the Social Security administrative cost of 0.6% on benefit returns paid out would reduce a 5.40% pre-expense return to a 5.37% return after expenses for those who start saving at age 45 (or 5.38% for those who start at age 22).  [The figures here come from a spreadsheet analysis by the author.]

The difference in returns after fees is huge.  A 2% fee on assets is close to 100 times the equivalent cost (in terms of return on assets) of the Social Security cost of 0.6% on benefits paid out.  Put another way, the 2% fee charged by the private fund managers reduces the return that would have been earned on equity investments by close to 40% (i.e. reducing a $100 return to only $60) while the Social Security charge would have reduced the return by only 0.5% (from $100 to $99.50).

The proposal set out here is to take advantage of this far greater efficiency of the Social Security system, by allowing Social Security to offer an optional retirement savings vehicle, into which Americans can invest a portion or even all of the investments now in their IRA, 401(k), or similar retirement savings plans.  The savings accrued in this way would then provide an individual supplement to your regular Social Security check once you retire.  Taking advantage of this would be completely voluntary.  The funds would be invested, like all current Social Security funds are now invested, in long-term US Treasury bonds.  Because of the inherent low costs in the Social Security system, the returns would be attractive.

C.  The Proposal:  A Supplemental Social Security Account

Most Americans (among those with any retirement plan coverage at their place of work) have seen their pension plans shift over recent decades from the traditional defined benefit plan to defined contribution schemes, mainly 401(k)’s.  From almost nothing prior to the 1980s, over 80% of private sector workers now who have any pension plan coverage, only have a defined contribution plan.  All workers may also have IRA’s, which are similar in magnitude (in terms of total assets) as what are held in 401(k)’s and similar plans, although the IRA assets primarily come from roll-over of 401(k) investments when a worker switches jobs.

In IRA’s, 401(k)’s, and similar defined contribution schemes, the workers must choose what investments to place their assets in.  There may be restrictions on the choices (e.g. many companies will limit the investments to certain mutual funds with whom they have contracted), but the aim is generally to provide a reasonable range of choices to the individual.  Unfortunately, all the choices come with high fees, which as discussed above, can severely reduce the investment returns.

The proposal here is to add the option of investing one’s IRA, 401(k), or similar plan assets into a supplemental Social Security account.  An individual would be allowed to choose to invest some, or even all, of their existing assets or any new savings set aside annually (up to the limits set by tax law for retirement savings that benefit from specified tax advantages) into such a supplemental Social Security account.  These funds would be invested, like all funds in the Social Security Trust Fund, in long-term US Treasury bonds.  Social Security would keep track of the contributions so made, and of the earnings from the interest on the US Treasury bonds (it would be a simple computer code).  Upon retirement and when the individual chooses to start to draw their regular Social Security benefit (normally at age 66 currently, although with this moving up to 67, and where there is some flexibility around these ages with the Social Security payments adjusted accordingly), Social Security would then increase the individual’s monthly Social Security benefit payment by an amount that would reflect the accumulated savings (including interest) in that supplemental Social Security account.  The accumulated savings would be converted into an actuarially fair annuity, reflecting the then expected life expectancy for an individual at that retirement age.

Note that the interest which would be paid into these accounts would come from the US Treasury.  But there would be no additional cost to the US Treasury since the bonds sold into these accounts would be offset by an equal number which would not then need to be sold to others, whether rich people who buy such bonds or the Chinese government, or whomever.

Set up on an actuarial fair basis, and with the relatively tiny Social Security administrative costs (of 0.6% of benefits paid) taken out, as for all Social Security payments, there would be no net cost to Social Security either.  And each individual (on an actuarial basis) would receive back by the time they die an amount equal to what they saved under this scheme, plus the interest earned on these assets.

But while each individual would end up even, there would be a significant amount invested in and supplementing the traditional Social Security Trust Fund at any point in time.  The Supplemental Social Security Fund would initially build up rapidly, as once the scheme is enacted, many people (all current workers) could begin to contribute to their new supplemental Social Security accounts, while only after such contributions have been made and with some time lag would there be retirees drawing down on their accounts.  We will present some projections below, and see that the new Supplemental Fund would eventually stabilize as a share of GDP, and would not decline.

But first we will discuss whether individuals would find investing into such Supplemental Social Security accounts attractive, and then discuss what the scale of the investments into such accounts might be, given the size of total retirement savings going each year into IRA’s and 401(k)’s.

D.  Investing Into Supplemental Social Security Accounts Would Be Attractive

It is of course impossible to predict what future returns will be.  But what can look at what the returns have been for periods in the past, and it is reasonable to conclude that for sufficiently long periods, the returns will likely be similar (or at least the relative returns across different asset classes will be similar, and that is what is relevant to an individual’s decision on asset allocations).

For the numbers here, I have looked at the 50 year returns for 1962 to 2012, using the data assembled by Professor Aswath Damodaran of NYU.  As noted above, adjusting for inflation (based on the CPI) over this fifty year period, an investment in the S&P 500 equity index would have yielded an annual real return of 5.4% (including dividends).  An investment in 10-year US Treasury bonds over this period would have yielded an annual real return of 2.6%.  And while it appears that data on actual returns people have achieved on average in their 401(k)’s or IRA’s do not exist (such data would be hard to collect, as most individuals do not even know what their actual returns, net of additions to their accounts, really were), one can calculate what an average 401(k) return might have been based on a weighted average of what returns might have been by asset classes held in these accounts currently.  Using Investment Company Institute estimates of the 2011 weights, and applying these weights to the 50 year returns by asset classes, the weighted average 401(k) return would have been 3.7% in real terms.

Subtracting what fees could be on such investments, the net returns would have been:

1962-2012 S&P500 Avg 401(k) 10-Year Treasury Bond
Gross Real Returns, before fees 5.4% 3.7% 2.6%
Private Return after fee of:
     2.65% 2.6%
     2.0% 3.3% 1.7% 0.6%
     1.5% 2.2% 1.1%
     1.0% 2.7% 1.6%
Social Security Return: 2.6%

With the very low costs of the efficient Social Security system, one would have earned a real return of 2.6% a year over this period, equal (within round-off) to the returns on the 10-Year Treasury bonds these funds would have been invested in.  The before fee return on equity investments would have been higher, at 5.4%, but with a 2% fee on assets incurred annually, the net return would have come down to 3.3%.  With a 2.65% fee (and some 401(k) plans cost more than even this), the return would have been 2.6%.  And while a 3.3% return is better than a 2.6% return, the equity return is much more volatile.  One’s returns will depend on the luck of the period during which one was invested, which will depend on the year you were born.  And there can be periods of a decade or more when the equity returns are even negative.  This has happened three times in the US since the 1920s:  during the 1930s in the Depression, in 1974/75 during Nixon/Ford, and following the 2008 economic and financial collapse at the end of the Bush administration.

Most people therefore do not hold solely equities in their 401(k)’s or IRA’s.  In addition to equities, they hold bond funds, mixed bond and equity funds, money market funds, as well as contracts guaranteeing some stable return (usually issued by life insurance companies, and with normally very high fees).  As discussed above, an estimate of what the weighted average return might have been yields 3.7% (based on 2011 weights, but 1962-2012 returns for the asset classes).  The estimate can only be approximate.  But based on it, the returns after fees would have been only 1.7% at a fee of 2.0%, 2.2% at a fee of 1.5%, and 2.7% at a fee of 1.0%.  Most 401(k) plans will charge fees substantially higher than 1.0% (when all the layers of fees are accounted for), but it is only at this low rate that one would get a return similar to what one would get by investing into a supplemental account at Social Security.

The return on a supplemental account at Social Security therefore looks attractive.  One should expect that if such an option were made available, Americans would choose to invest a significant portion of their retirement assets into such an account, and possibly even all of their retirement assets.  The returns would be secure, relatively stable, and simple to manage.

E.  What Might Be the Level of Such Investments in the Aggregate?

The next issue to address is how much such investments might add up to.  One needs this in order to calculate what the impact might be on the size of the Social Security Trust Fund.

The truth is that one does not really know.  An option might be attractive, but few then sign up, or many do.  The numbers that will follow here are therefore speculative.  But they will provide some sense of the magnitudes involved.

Due to tax and other reporting requirements, there are good numbers on the stock of assets in IRA’s, 401(k)’s, and similar defined contribution (and other) pension plans.  These figures largely come from the US Federal Reserve Board Flow of Funds estimates, but are buried there with much other data, and a more convenient source of the retirement plan assets are provided by the Investment Company Institute.  As of the end of 2012, the ICI estimates total assets held in IRA’s came to $5.4 trillion, and total assets in defined contribution plans (mainly 401(k)’s, but also others) came to $5.1 trillion.  These two categories together thus came to $10.5 trillion of assets, 67% of 2012 GDP, that are invested according to the directions of individuals.   Total pension fund assets, primarily in defined contribution plans (including plans for government workers) and annuities, but excluding Social Security, came to $19.5 trillion as of the end of 2012, or 124% of 2012 GDP.

But while good figures are available on total assets in these plans, they are more sparse on annual flows into the plans.  Figures are available on IRA’s, where over the ten year period 1998 to 2008 (where 2008 is the most recent year available) gross new flows, including new contributions and from roll-overs, averaged a bit over $280 billion per year in the ICI numbers.  Assuming similar flows into 401(k)’s and other such defined contribution plans, but scaled to reflect their slightly smaller size ($5.1 trillion vs. $5.4 trillion), it would be reasonable to assume about $270 billion is flowing annually into such plans.  The two together would then be $550 billion a year, or about 3.5% of GDP.

One can only then guess what share of such flows might be attracted into the Supplemental Social Security accounts being discussed here.  With roughly a third of 401(k) assets in bond funds, and a further 9% in short term investments (mainly money market funds), it might be reasonable to assume that perhaps 30% of new flows into IRA’s 401(k)’s, and similar plans, would be attracted to Supplemental Social Security accounts.  Such accounts would be similar in nature to bond and money market investments, but with significantly greater returns due to the far lower investment fees of Social Security, combined with lower risk and greater security as these investments would be in US Treasuries.

While the 30% share is a guess, it is a reasonable bench-mark.  The actual flows could turn out to be significantly higher (particularly as an attractive new low-cost retirement savings option such as this might well lead to higher savings going into IRA’s and 401(k)’s), but it could also be lower.  A 30% share of $550 billion in annual flows would come to a little over 1% of GDP.

F.  Impact on the Social Security Trust Fund

Rounding down, I calculated what the impact would be assuming initial flows equal to 1% of GDP, and with the new retirement savings option entering into effect in 2015.  One then also needs, for the period of the projection (now to 2086), estimates of GDP growth, population growth, and real wage growth.  I used the projections made by the CBO for their long term Social Security projections, cited above, where over this period real GDP would grow at an annual average rate of 2.25%, and population would grow at an annual average rate of 0.58%.  I assumed real wages would on average grow at the rate of per capita GDP, or 1.66% annually.  Real wages have grown at a slower rate than per capita GDP since the 1980s and the Reagan “revolution” (as was presented and discussed in this earlier post on this blog), but it is assumed this cannot go on forever.   Finally one needs for the projections an assumption on the return to investments in long-term US Treasuries.  For this I assumed that the returns going forward would match the 2.6% real return of the 50 year period 1962-2012.

The spreadsheet then calculated the savings going into the Supplemental Social Security accounts of each age group, changing over time based on real wage growth and population growth.  I assumed workers would retire at 67, begin work at 22, and die at 92 (on average), with the savings converted into an annuity upon retirement at 67 with level payments (but reflecting accumulation at the real rate of interest of 2.6% on the assets in the accounts).  Note that the life expectancy of an assumed 92 is the life expectancy of someone who has reached the age of 67, and not the life expectancy at birth.  And while this life expectancy is less than 92 now, the projections are very long term, and were kept simple in part by not including variable life expectancies.

Each individual would then accumulate assets in the accounts, from savings plus the interest earned, until age 67, and would then draw down the accumulated assets through an actuarially fair annuity which would pay out a fixed amount (in real terms) each year until they die (which on average was assumed to be age 92).  For each individual, the amount paid in, including accrued interest, would then match the amount paid out.  But there would be a positive balance in the Supplemental Social Security Trust Fund accounts at any point in time, which would supplement the traditional Social Security Trust Fund.  The traditional Social Security Trust Fund would remain unchanged.

The resulting path of the new Supplemental Fund would be as follows:

Social Security Trust Fund, Traditional & Supplement, 2015-2086

The Supplemental Fund would at first grow, both in real terms and as a share of GDP, as initially all the age groups would be paying into the accounts while no one would be taking out funds until time had passed.  The amounts being taken out would then grow over time.  With the constant growth rates being assumed for GDP, real wage growth, and population growth, and the return of 2.6% in real terms on the investments, it is perhaps not surprising that the Supplemental Fund rises as a share of GDP and then flattens out.  From about 2055 onwards (40 years from the start), it would remain constant at about 24% of GDP.  And it is important to note that the Supplemental Fund would not decline as a share of GDP, but remain flat.

The traditional Social Security Trust Fund is projected to decline, due to the annual negative net cash flows shown in the graph at the top of this post, and will reach zero in 2035 according to the CBO projections.  It would continue to fall, and at an increasing rate over time (due to interest on what would then be negative assets) if nothing is done to make up for the annual shortfall.  The Supplemental Fund would not affect this path.  However, the Supplemental Fund would provide funding to the Social Security system, and when combined with the traditional Social Security Trust Fund, the overall fund would not now fall to zero until about 2058.  That is, it would extend the life of the Social Security Trust Fund by about a generation (23 years) before Social Security taxes or some other taxes would need to be raised to provide the scheduled benefits to Social Security recipients.

G.  Conclusion

The Supplemental Social Security accounts would be completely voluntary and would present an attractive option for retirement savings due to the low cost and high efficiency of the Social Security system.  It would also provide funding to the Social Security system which would extend the life of the Social Security Trust Fund, before something would need to be done, by a generation.  And the Supplemental Fund would not build up and then come down, putting additional stress on the Social Security system, but would rather build up to about 24% of GDP (under the assumptions made above) and then stay there.

The Supplemental Social Security accounts would also not be a Ponzi scheme.  Each individual would get back (on an actuarial basis) exactly what they put in, with interest.  For the economy as a whole, the fund would rise (to 24% of GDP under the particular assumptions made on growth) and then stay there.  And the interest being paid would come from the US Treasury.  Since the US Treasury would be paying the interest to others anyway, the shift to the Supplemental Social Security fund would not present any additional burden on the US Treasury, but would rather be neutral.

But it needs also to be recognized that a scheme such as this would not represent a permanent fix to the traditional Social Security system.  It would provide significant funding to Social Security, which would extend the life of the Trust Fund, but would not solve the underlying deficit that has arisen due to longer life expectancies with an unchanged (since 1990) Social Security payroll tax rate.  As noted above, the gap is not large, at about 1% of GDP, and is far less than the tax revenues lost as a result of the Bush tax cuts.  But perhaps by 2058 the country will be mature enough to recognize that the most successful social program of the last century in the US is worth paying for.

The Ryan Budget Plan: Simply Not Serious

Posted on by
Ryan Budget Plan Actual Projected Change from 2011 
% of GDP 2011 2023 2030 2023-2011 2030-2011
Total Spending excluding Interest 22.50 17.25 17.50   -5.25   -5.00
A. Medicare 3.25 3.50 4.25     0.25     1.00
B. Medicaid & CHIP 2.00 1.25 1.25   -0.75   -0.75
C. Social Security 4.75 5.50 6.00     0.75     1.25
D. All Other 12.50 6.75 5.75   -5.75   -6.75
  1) Defense including VA 5.75 4.50 4.50   -1.25   -1.25
  2) Rest of Govt (calculated) 6.75 2.25 1.25   -4.50   -5.50
Sources:  For Defense spending line:  CBO March 2012 “Updated Budget Projections”.  Assumes constant Defense (including VA) spending in real terms to 2023, with no sequestration.  The Ryan Plan calls for more, but it is not clear how much more, so this is a minimum.  Assumes constant share of GDP after that.
For all other lines:  CBO March 2012 evaluation of Ryan Budget proposal.  The CBO numbers are presented to the nearest quarter of a percentage point of GDP, and is shown here in decimals simply for clarity.  Note:  Totals may not add due to round-off in the CBO figures.

With Mitt Romney’s choice of Congressman Paul Ryan as his vice-presidential running mate, the policy positions that Ryan has taken over the years have become of more interest.  As Chairman of the House Budget Committee, Ryan is particularly well known for the plans he put forward each spring for the federal budget.  Even ostensibly neutral news reporters have praised these plans as serious proposals to remedy America’s fiscal ills, noting that while one might not agree with certain of the specifics, the plans at least set forth a serious and internally consistent set of proposals to bring down the deficits.

Superficially, the plans might convey such an appearance.  But as soon as one starts to dig into the numbers, one finds major issues.  The Ryan Plan does not really reduce the deficit compared to a scenario where the Bush Tax Cuts are allowed to expire, but rather slashes spending in order to bring down the deficit to what it would be without the Bush tax cuts.  Ryan then goes further by proposing even higher tax cuts on top of this.  Defense spending would be increased; there are specific proposals to cap spending on Medicare and on Medicaid; and Social Security (in the 2012 proposal) is left as is.  Everything else government spends funds on, from assistance to the poor, to NASA, to border security, and to the Judiciary, would be drastically slashed.  Ryan does not present detail on which programs would be cut the most, but only some vague commentary, but these programs as a whole would be slashed by at least two-thirds by 2023 (as a share of GDP), by over 80% by 2030, and by essentially 100% by 2040.  After that, there would have to be negative spending on such programs for Ryan’s budget to add up.  This is simply not serious.

The rest of this blog post will document and discuss these points, and others.  A note on sources:  The Ryan Budget Plan is as presented in his budget document of March 2012.  The CBO at that time presented its own analysis of the budgetary implications of the Ryan Budget Plan, as is standard and as was requested by Ryan.  For other CBO estimates, I have taken figures from the CBO projections of March 2012, to coincide in time with when the Ryan Plan (and the CBO analysis of it) came out.  The CBO has updated its standard projections since then, but the changes have not been significant, in particular for the numbers a decade out or more.  For consistency and to see where the differences are, it is therefore best to use the March CBO numbers when making comparisons to those that would result from the Ryan proposals.

The main points are:

1)  The budget deficit under the Ryan Plan is not significantly lower than it would be compared to simply allowing the Bush Tax Cuts to expire.

Ryan presents his Budget Plan as perhaps unfortunately tough on many, but nonetheless critically necessary in order to keep the government debt from exploding.  He starts his presentation with scary figures, showing an ever rising government debt to absurd levels if the “current path” is followed.  He asserts his plan is necessary to bring this under control.

But what Ryan does not show is the path that would be followed if the Bush tax cuts are allowed to expire.  The Bush tax cuts were never permanent, but rather have always been scheduled to end.  The CBO presents such a path in their “Baseline” and “Extended Baseline” scenarios.  The federal deficit under the Ryan Plan would be 1 1/4 % of GDP in 2023, while under the CBO Extended Baseline Scenario it would be 1 3/4% of GDP.  Both of these numbers are modest, and the difference is well within the forecasting error one would expect for a figure eleven years in the future.  It should be noted that the CBO Extended Baseline scenario also includes some federal spending cuts relative to what they would be under current policy, but these cuts are not nearly as drastic as the cuts under the Ryan Plan, plus they are relatively small (accounting for only one-fifth to one-quarter of the impact of allowing the Bush tax cuts to expire).

The Bush tax cuts are currently scheduled to expire on December 31, 2012, and I would certainly not advocate allowing them to expire abruptly on that date.  Unemployment is still high due to a lack of demand for what the economy could produce, and in a pure Keynesian fashion higher taxes on consumers for a given level of government expenditure will be depressive.  Rather, the tax cuts should be phased out over a period of a few years (and start with phasing out the beneficial rates for the very rich, as Obama has proposed).  The long term benefits of deficit reduction would still come through, but taxes on those whose consumption would be impacted would not increase while unemployment is high.

It is therefore simply not true that the Ryan Budget Plan is necessary to bring down the federal deficit to sustainable levels.  The CBO shows that under current law, which provides that the Bush tax cuts would be allowed to expire, the deficit would come down similarly over the next decade.  This point was made in an earlier posting on this blog on the CBO projections that were issued in June (see here), and in an analysis focused more specifically on just the Bush tax cuts from last February (in this post).

2)  The deficit does not fall significantly despite a drastic slashing of much of government spending, since Ryan would not only make the Bush tax cuts permanent, but would cut taxes even further by a similar amount.

Extending the Bush tax cuts would cut tax revenues by 2 3/4% of GDP in 2023, by 4% of GDP in 2030, and by more later according to the CBO projections.  (This can be worked out from the difference in revenues between the CBO Baseline Scenario and its Alternative Fiscal Scenario, where both are presented in the CBO analysis of the Ryan Plan.)  The very long term scenarios should not be taken too seriously as so much will certainly change by then, but they give an indication of the trends implied by the scenarios.  Ryan would then add further tax cuts on top of the Bush tax cuts, reducing the top marginal tax rate to 25% and reducing the progressivity in the tax system by consolidating all taxes into just two brackets of 10% and 25%.  He would also cut the top tax rate on corporate profits to 25% from the current 35%.

The additional tax cuts proposed by Ryan would cut federal tax revenue by a further 3% of GDP by 2023 if nothing additional is done (based on an analysis of the Ryan Plan by the non-partisan Tax Policy Center, and comparing their results to the CBO Baseline Scenario).  This is similar in magnitude to the losses in tax revenues resulting from the Bush tax cuts.

3)  Ryan does, however, propose to offset the additional 3% of GDP of revenue losses resulting from his additional tax cuts, by reducing or eliminating certain (but unspecified) tax deductions or tax preference items.  But it is impossible to do this without leading to higher taxes on the poor and middle classes, to offset the tax cuts going to the very rich.

Ryan asserts he would fully pay for his additional tax cuts, making this part of the plan revenue neutral, by eliminating or reducing certain tax deductions and tax preference items.  But he refuses to say which deductions and other preferences would be reduced or eliminated.  He does say that the highly preferential tax rate on capital gains and dividends of just 15% would not be raised (and in his 2010 budget plan he would have brought this rate all the way down to zero).  This preferential tax rate on income received from wealth of course benefits the wealthy the most.

Especially with the wealthy able to continue to enjoy the benefits of this preferential tax rate on capital gains and dividends, it is impossible to make such a tax plan revenue neutral without raising taxes on the majority of Americans.  The non-partisan Tax Policy Center did an analysis of the implications in the broadly similar Romney tax plan.  While there are differences in the detail between Romney and Ryan, both would cut marginal tax rates sharply, especially for the rich, and both assert they would then reduce or eliminate certain deductions and other tax preferences (but not the capital gains tax preference) to raise sufficient revenue to make it overall revenue neutral.  And both keep secret which deductions or other tax preferences they would reduce or eliminate.  The Tax Policy Center therefore took the most extreme case possible, by assuming that all the remaining deductions and tax preferences of the highest income group would be cut first (and cut in full for them), followed by that for the next highest income group, and so on down the scale until enough had been raised by cuts in deductions and tax preferences so as to offset the revenue losses from lower tax rates.

The Tax Policy Center found that even in this most extreme case for the Romney tax plan, taxes due from the richest 5% of the population would still fall (and some very substantially:  i.e. fall by $87,000 for those making over $1 million in income), while taxes would have to rise for remaining 95% of the population.  And this is the most extreme possible case.  In any real program, it is possible that only the very richest 2 or 3 or 4% of the population would see a net decrease in taxes (as they would still enjoy at least some deductions and tax preferences) while the remainder of the population would see a net increase in their taxes.

The Ryan tax plan, while not yet analyzed by the Tax Policy Center, would have a similar effect.  That is, the very richest in the population would gain, while the over-whelming majority of the population would lose.

It is therefore not surprising why Ryan would want to keep secret which deductions and tax preference items he would cut.  But if he insists his tax plan would be revenue neutral, it is mathematically impossible to cut taxes for some without raising them for others.  And when the tax cuts benefit the rich by a disproportionate amount, as they do under these plans to cut the top marginal tax rates, a disproportionate number of the poor and the middle class will need to see their taxes rise to make up for such tax losses.

4)  While slashing much of government spending in order to keep the deficit under control while slashing taxes, Ryan insists on increasing defense spending.

Ryan would not slash all government spending.  Indeed, he would increase spending on defense beyond what Obama would have.  While he provides some figures in his plan, it is difficult to relate them to the CBO numbers, as the base used for each differs.  He does say (on page 21 in his plan) that his budget “ensures that the defense budget grows in real terms in each year” over the ten-year period covered by the budget resolution (2013-22).  But it is not clear how big of a real increase he would have.  The line on defense spending in the table at the top of this blog therefore takes the CBO figures for what defense spending (including for the Veterans Administration) would need to be to be constant in real terms until 2023.  Ryan would spend more than this on defense, but it is not clear how much more.  After 2023, defense spending is assumed then to remain steady as a share of GDP.

The line in the table on defense spending should therefore be viewed as a minimum, where Ryan would spend more.  This is important, since the line in the table on “Rest of Government” would be squeezed even more if defense spending is higher.

5)  While asserting loudly that Social Security is on an unsustainable course, the Ryan budget plan does not propose any changes in Social Security.

Rather, all Ryan proposes in his budget is that the President submit a plan to address this.  The Social Security expenditure figures in the budget are the same as in the President’s budget, and are what the Social Security Board of Trustees project will be required.

Earlier in his career, Ryan had proposed diverting what is paid in Social Security taxes to individual private accounts.  As had been noted in an earlier entry on this blog, such a plan would, if fully implemented, divert tax revenues equal to 5% of GDP to privately managed accounts, thus increasing the fiscal deficit by 5% of GDP unless other taxes are raised.  These privately managed accounts would also be enormously more costly to individuals than the current program is, as private fund managers charge fees an order of magnitude greater than the administrative costs of the Social Security Administration.  And they would expose individuals to market risks, which as we have seen in recent years many individual investors (in managing their IRAs, 401k’s and similar pension plans) are not in a good position to handle.

6)  While Ryan proposes to end Medicare as a program that ensures medical coverage for all seniors, his proposals would be phased in (applying only to those age 55 and below) and would not have a major impact on government spending for several decades.

Paul Ryan is perhaps best known for his proposal to end Medicare as a program that ensures medical coverage for all seniors, and replace it with a voucher scheme where seniors would instead be given a limited coupon, which could be used to pay part of the cost of private medical insurance.  There are many problems with such a scheme, which merits a separate blog post by itself.  But from the point of view of the budget, the primary point is that Ryan would limit the size of the individual vouchers to some fixed amount, rather than have Medicare insurance cover the cost of medical care, whatever that cost turns out to be.  That is, he shifts onto seniors the burden of any costs above whatever amount he believes the federal budget should cover (given whatever tax cuts are implemented).  Rising medical costs certainly are a problem, but Ryan does not address this.  Rather, he would simply shift the burden of rising costs onto seniors, and for those not then rich enough to afford coverage like they have now, he would say too bad.

This Ryan plan was strongly criticized by many when it came out.  In a more recent variant, Ryan says he would retain traditional Medicare as an option.  However, such a plan would not be sustainable.  Budgeted medical costs would still be capped, and with traditional Medicare obligated to accept any senior who applies for this option, those most in need for medical care would choose Medicare coverage, while private insurers would seek to enroll (whether overtly, or more subtly through the design of their coverage) the more healthy and hence less costly patients.  Traditional Medicare would then soon be bankrupted as it took on the burden of the seniors with the highest medical expenses.

Ryan’s proposals have not been popular.  This has been especially clear for those over age 65 who are now receiving Medicare coverage, and for those approaching 65 who will soon depend on Medicare.  They naturally have been the ones to follow the issue most closely.  For solely political reasons, Ryan therefore proposed that his changes to Medicare would only apply to those currently under the age of 55.  If Ryan’s proposals were in fact of benefit, there is no reason why they could not be implemented immediately.

Because Ryan’s proposal to end Medicare as an insurance scheme would only apply to those currently below age 55, there will be no impact on the budget for ten years.  It would then impact the budget only slowly as the new seniors phase into the different plan (the existing Medicare recipients would continue to participate in the traditional insurance program).  There would then be no budgetary impact, relative to the CBO Baseline, for the next ten years.  And this is indeed what he assumes, with Medicare spending then still rising as a share of GDP relative to 2011 as an increasing share of the population (the baby boomers) turn 65 and enroll in Medicare (and as shown in the table at the top of this blog).

But Ryan then made what must have been an overlooked error.  As was discussed in an earlier posting on this blog, Ryan assumes in his budget the same $716 billion in savings (over ten years) in Medicare expenses resulting from the implementation of Obamacare, as the Obama budget does.  This savings to what Medicare would otherwise have to pay hospitals and other health care facilities arises because of the reduction in the number of uninsured under Obamacare, with consequent cost savings to these hospitals and other facilities.  Hospitals currently can recover only a part of their costs from treating the uninsured, and hence must shift these costs onto those covered by Medicare or by private insurance.  With Obamacare and fewer uninsured, such cost shifting will be reduced.  The estimated savings is $716 billion.

But Ryan (and Romney) would end Obamacare.  Ryan’s budget assumes Obamacare would be immediately reversed.  But the hospitals and other health care facilities would not then realize the $716 billion in savings from fewer uninsured to treat.  However, Ryan assumes that Medicare would still see a savings of $716 billion in what it would pay out.  This is a blatant inconsistency.  Ryan is treated as a good numbers person, but a good numbers person would not make such a simple mistake.  And if not a mistake, it is fraud.  He (along with Romney) also continues to call the $716 billion in Medicare cost savings as a “raid” on the Medicare Trust Fund (see again the earlier blog posting referenced above), when it is the opposite.

7)  In contrast to Medicare, the Ryan budget would immediately slash spending on Medicaid and CHIP.

While much of the discussion has been on Ryan’s proposal to end Medicare in its current form, the big cuts in medical care in Ryan’s proposed budget would be to Medicaid (medical insurance for the poor, including a portion of costs for the elderly poor) and in CHIP (the Children’s Health Insurance Program).  The CBO Baseline forecasts that spending on these two programs, to maintain current standards, would need to rise from 2% of GDP in 2011, to 3% of GDP in 2023 and 3 1/4% of GDP in 2030.  The increase would be due to a changing demographic structure among the poor, as they get older on average (as the overall population does) and hence require more in medical care.

Ryan, in contrast, would slash federal spending on Medicaid and CHIP  to 1 1/4% of GDP in 2023 and still 1 1/4% of GDP in 2030.  This would be a cut of close to 60% in 2023 and over 60% by 2030.  Ryan justifies such cuts by saying he would transfer the capped funds as “block grants” to the states.  Block grants give the states flexibility in how they could then spend these funds.  Ryan argues that the states could then become more efficient in delivering these health care services.  But it is nonsense to believe that the program funds could be cut by 60% and still achieve anything close to current coverage.  The poor would clearly suffer.

8)  All other federal spending would be slashed under the Ryan budget, to levels that are simply not credible.

As shown in the table at the top of this blog, Ryan would cut total federal spending (excluding interest) from 22 1/2% of GDP in 2011 to 17 1/4% of GDP in 2023 and about the same in 2030.  While not included in the table above, the CBO assessment of the Ryan plan projected that such spending under his proposals would then fall further to just 16 3/4% of GDP in 2040 and to 15 1/2% of GDP in 2050.

To arrive at these totals, and given the spending on Medicare, Medicaid and CHIP, and on Social Security, the total spending on defense and on everything else in the government budget would fall under Ryan’s proposals from 12 1/2% of GDP in 2011 to just 6 3/4% in 2023 and 5 3/4% in 2030.  Although not shown above, it would fall even further to 4 3/4% of GDP in 2040 and to 3 3/4% in 2050.

The CBO noted that spending on this category had never fallen below 8% of GDP in any year since before World War II.  The figures Ryan proposes are simply not credible.  But it gets even worse.  As noted above, Ryan says specifically that he would not cut Defense spending, but rather have it increase in real terms over at least the next ten years.  He was not clear by how big an increase in real terms it would be, but it would be an increase.  The table above assumes that spending on defense (including for veterans) would be constant in real terms until 2023, and then constant as a share of GDP thereafter.  This is conservative, and Ryan would in fact spend more on defense, although it is not clear how much more.

Even with this conservative assumption on what Ryan intends for defense spending, the spending on the rest of government (other than on defense, major medical programs, and Social Security) would fall from 6 3/4% of GDP in 2011 to just 2 1/4% of GDP in 2023.  This is a cut of two-thirds on everything else the government does.  And on these assumptions, such spending would fall further to essentially zero in 2040 and to a negative amount in 2050.  It is impossible to “spend” negative amounts.

Conclusion

Ryan’s budget proposals simply do not add up to a credible program.  There are internal inconsistencies; he projects cuts in what government spends other than on defense, Social Security, and major medical programs, that are simply not credible and ultimately mathematically impossible; and he leaves out key specifics such as what tax deductions and tax preferences would be cut to make up for lost revenue, and what specific government programs would be cut to lead to the totals he proposes.

Furthermore, the bottom line deficits following from his proposals would not be that different from the deficits in the CBO Baseline over the next decade (and is only cut by more in the long term by his assumption that the residual government programs can be cut to levels that are simply not credible, and ultimately to levels implying negative spending).

This is not a serious program.

Health Insurance Coverage: Imposition of Religious Beliefs is Not Religious Freedom

Posted on by

Republicans are attacking the Obama administration, in often vicious words, with the charge that Obama is leading an “assault on religion” (Romney), that he “has declared war on religious freedom in America” (Gingrich), and that “the president has reached a new low in this country’s history of oppressing religious freedom that we have never seen before” (Santorum).  And yesterday, in the nationally televised Sunday morning talk show “This Week” on ABC, Santorum extended his attack to the late President John F. Kennedy, saying that the well known speech during the 1960 campaign to the Greater Houston Ministerial Association, in which Kennedy stated his belief in separation of Church and State, made him almost “throw up”.

These attacks are appalling.  They originated in a dispute on what should be included in health insurance plans for those plans to be considered as minimally adequate in terms of what they cover, and not a sham.  Understanding why this is important is central to understanding the health insurance reforms identified with Obama.  This will be reviewed here.  And once this is understood, it will be clear that to permit religious institutions, or indeed anyone else, to impose their particular religious beliefs on their employees or on others over whom they have authority or leverage, is not religious freedom but rather its antithesis.

The aim of the Obama health insurance reforms is to allow access by all Americans to health care services.  This could have been achieved in a variety of ways (as we see from what other countries have been able to do at far lower cost than the system in the US), but the realities of the politics in Congress dictated that this had to be done through employer based health care plans and private health insurers, which already cover most but not all Americans.  But for such universal coverage to have any meaning, one has to define what a minimally adequate health insurance plan would cover for the employees.

The dispute arose over whether minimally adequate health insurance plans would cover the costs of contraceptive services and certain other health services for women.  No employee would be forced to use such services:  that would be up to them.  The question is whether religious and other institutions should be allowed to prohibit their employees from having access to such services through their health care plan, based on a religious belief that such services should not be used.  The Catholic Church was the most prominent opponent to allowing such access, which it wanted to extend not only to priests and nuns, but also to all other employees (such as nurses or janitors) of institutions such as universities or hospitals they operate or even direct business operations (such as in publishing).

It is worth noting at this point that this would not have been an issue if the health insurance reform had allowed for a “public option” health insurer, which would compete with private health insurers and would be available to all.  This was killed by Joe Lieberman, Senator for Connecticut, a state where several prominent private health insurers are headquartered.  With a public option insurer available, one could ensure a minimally adequate set of services was available to all without the need for doing this through private insurance plans.  But private insurers did not want to face such competition.

But without such a public option to serve as an alternative, the private health insurance plans need to meet some minimum standard for them to be meaningful.  The question is whether the Catholic Church or any other employer should be allowed to prohibit access through the employee health insurance plans to certain health care services that they believe run counter to the religious beliefs of the employer.  They argue that “religious freedom” means they have the right to impose particular religious beliefs on their employees.

But this is not religious freedom, but the opposite.  Religious freedom is the right to hold the religious beliefs that you do, not the right to impose certain religious beliefs on others.  If certain religious views can be imposed, by organized religions directly, or by the state at the behest of organized religions, one does not have religious freedom, but rather Iran.

And if the heads of religious organizations can block access by their employees to certain health care services based on religious views, where would this stop?  The current debate is over whether women would have access to contraceptives.  Santorum has argued women should also be denied access to pre-natal testing.  In the past, there have been religious leaders who have wanted to limit access to patients needing treatment for syphilis and gonorrhea, or for HIV/AIDs, saying these diseases reflected God’s just punishment for those who have sinned.  Would religious organizations similarly be allowed to limit access to treatments resulting from smoking tobacco, or from too much alcohol, or from obesity?

There is also the apparent confusion by the Republican candidates and others that for some reason the health insurance plans provided to the employees are not part of the wage compensation package provided to the employees for the work that they do.  Rather, they see them as a “gift” or something similar from the employers.  But they are not “gifts”.  They are part of the compensation package the employee earns in return for the work that he or she does, just like wages paid directly.

And if religious institutions would now be allowed to restrict access through one part of the employee compensation package (the health insurance plans) to certain expenditures based on religious beliefs, would there be any limits?  Suppose employee wages started to be paid through Debit cards or some similar technology to track what the wages were used for.  Would the religious institutions then be permitted to prohibit their employees from using their wages to purchase birth control pills or condoms on their own?  And one can then quickly go to even more absurd examples, such as Catholic institutions prohibiting their employees from purchasing meats in restaurants on Fridays during Lent, or Jewish or Muslim institutions not allowing their employees to buy pork, or Mormon or other religious groups not allowing their employees to buy caffeinated beverages such as coffee or Coca-Cola, and no one being allowed to buy a subscription to Playboy?

On Sunday, Santorum extended this to an even more worrying belief.  On “This Week” on ABC (for a transcript, see here), Santorum said that on reading a transcript of President Kennedy’s famous speech to the Greater Houston Ministerial Association during the 1960 campaign (for a transcript and video, see here), he “almost threw up”.  When asked to clarify, Santorum made clear that he is opposed to what most of us thought was the now well accepted Constitutional principle of separation of church and state, as stated in Amendment 1 to the Constitution.  The interviewer (George Stephanopoulos) asked “Why did it [the Kennedy speech] make you throw up?”  Santorum responded “Because the first line, the first substantive line in the speech says, ‘I believe in an America where the separation of church and state is absolute.’  I don’t believe in an America where the separation of church and state is absolute.”  Santorum argues that those with particular religious beliefs should be allowed, through their representatives in the government, to impose their particular religious views on all Americans.  A minority would no longer have Constitutional protections from being forced to adhere to such religious beliefs.  His argument is a confused one, as he argues that with separation of church and state, those with religious beliefs are somehow excluded from public debate and that non-believers then impose their views on all.  He therefore wants the opposite, ignoring the fundamental point that no one should be allowed to impose particular religious (or non-religious) views on anyone.

This would then not be religious freedom, but rather religious tyranny.