The Sluggish Recovery: Fiscal Drag Continues to Hold Back the Economy

Recessions - GDP Around Peak, 12Q before to 22Q after

I.  Introduction

The recovery from the 2008 economic collapse remains sluggish, with GDP growing in the first half of 2013 at an annualized rate of only 1.4% (according to recently released BEA estimates).  And based on fourth quarter to fourth quarter figures, GDP grew by only 2.0% in 2011 followed by just 2.0% again in 2012.  As a result, the unemployment rate has come down only slowly, from a peak of 10.0% in 2009 to a still high 7.4% as of July.

Conservatives have asserted that the recovery has been slow due to huge and unprecedented increases in government spending during Obama’s term, and that the answer should therefore be to cut that spending.  But as has been noted in earlier posts on this blog, direct government spending during Obama’s term has instead been falling.  This reduction in demand for what the economy can produce has slowed the recovery from what it would have been.

This blog will update numbers first presented in a March 2012 post on this blog, which compared the paths of GDP, government spending, and other items in the periods before and after the start of each of the recessions the US has faced since the 1970s.  That earlier blog post looked at the paths of GDP and the other items from 12 quarters before the business cycle peak (as dated by the NBER, the entity that organizes a panel of experts to date economic downturns) to 16 quarters after those peaks (when the downturns by definition begin).  The figures were rebased to equal 100.0 at the business cycle peaks.  We now have an additional year and a half of GDP account data, so it is now possible to extend the paths to 22 quarters from the start of the recent downturn in December 2007.  This has therefore been done for all.

The conclusions from the earlier post unfortunately remain, but are even more clear with the additional year and a half of observations.  GDP growth remains sluggish, government spending has fallen by even more, and residential investment remains depressed (although it has finally begun to recover).

II.  The Path of Real GDP

The graph at the top of this post shows the path followed by real GDP in the periods from 12 quarters before to 22 quarters after the onset dates of each of the recessions the US has faced since the 1970s.  The sluggish recovery from the current downturn is clear.

The economy fell sharply in the final year of the Bush administration, and then stabilized quickly after Obama took office.  GDP then began to grow from the third quarter of 2009 and has continued to grow since.  But the pace of recovery has been slow.  By 22 quarters from the previous business cycle peak, real GDP in the current downturn is only 4% above where it had been at that peak.  At the same point in the other downturns since the 1970s, real GDP was between 15% and 20% above where it had been at the previous peak.  This has been a terrible recovery.

 III.  The Path of Government Spending

How has this recovery differed from the others?  To start to understand this, look at the path government spending has taken:

Recessions - Govt Cons + Inv Expenditures Around Peak, 12Q before to 22Q after

Direct government spending has fallen in this recovery, in sharp contrast to the increases seen in the other recoveries.  Real government spending was 26% higher by 22 quarters after the onset of the July 1981 recession (the green line) during the Reagan presidency, and 13% higher at the same point in the recovery from the March 2001 recession (the plum colored line) during the Bush II presidency.  While both Reagan and Bush claimed to represent small government conservatism, government spending instead rose sharply during their terms.

In contrast, in the current downturn direct government spending is now 1.2% below what it had been at the start of the recession in December 2007.  Furthermore, it is worth noting that while it rose in the final year of the Bush presidency and then in the first half year after Obama took office (a major reason why the recovery then began), it has since fallen sharply.  Government spending is now almost 7% below where it had been in mid-2009, a half year after Obama took office.  Such a decline (indeed no decline) has ever happened before, going back at least four decades, as the economy has struggled to recover from a recession.  The closest was during the Clinton years, when government spending was essentially flat (a 1% increase at the same point in the recovery).

Note that the measure of government spending shown here is that for total government spending on consumption and investment (i.e. all government spending on goods and services).  This is the direct component of GDP.  Government spending can also be measured by including transfer payments to households (such as for Social Security or unemployment insurance), but as was noted in the earlier blog post from March 2012, the results are similar.  Note also that the government spending figures include spending at the state and local levels, in addition to federal spending.  While we speak of government spending as taking place during some presidential term in office, the decisions are made not simply by the president but also by many others (including state and local officials, and the Congress) in the US system.  But the president at the time is typically assigned the blame (or the credit) for the outcome.

IV.  The Path of Residential Investment

The current downturn and recovery also differs from the others by the scale of the housing collapse, and consequent fall in residential investment:

Recessions - Residential Investment Around Peaks, 12Q before to 22Q after

The build up of the housing bubble from 2002 to 2006 was unprecedented in the US, and the collapse then more severe.  As the graph above shows, there has been a start in the recovery of residential investment from the lows it had reached in 2009 / 2010, but it is still far below the levels seen in previous downturns.

Housing had been overbuilt during the bubble in the Bush years, leaving an oversupply of housing once the bubble burst.  And while supply was in excess, demand for housing was reduced due to the severe recession.  As was discussed in an earlier blog post on the housing crisis, the result was a doubling up of households as well as delays in household formation as young adults continued to live with their parents.  Residential investment therefore collapsed, and has recovered only very slowly.

V.  The Path of Household Debt

The housing bubble also led to over-indebtedness of households.  Nothing of this sort at all close to this scale had ever happened before in the US.  With the lack of regulation and oversight of the financial sector during the Bush administration, banks and other financial entities launched and aggressively marketed and sold financial instruments that led to a bidding up of home prices.  But these new financial instruments were only viable if housing prices continued to rise forever.  When the housing bubble burst, widespread defaults followed.  And those households who did not default struggled to pay down the debts they had taken on, for assets now worth less than the size of the debts tied to them.

The result was a sustained fall in household debt (three-quarters of which is mortgage debt) in the period of the downturn:

Recessions - HH Debt Around Peak, 12Q before to 22Q after

This pay-down of debt had never happened before, and is in stark contrast to the rise in household debt seen in all the other downturns of the last four decades.

VI.  The Path of Personal Consumption Expenditure

Households struggling to pay down their debt have to cut back on their consumption expenditures.  This brings us to the last element of the current recovery I would like to highlight:  the especially slow recovery in household consumption.  That path of consumption during the current downturn stands out again in contrast to the paths followed in the other downturns and recoveries of the last four decades in the US:

Recessions - Personal Consumption Around Peaks, 12Q before to 22Q after

The difference is stark.  Households could spend more in the prior recoveries in part because they could continue to borrow (see the graph on household debt above).  In this recovery, households have instead had to pay down the debts they had accumulated in the housing bubble years, and could increase their household consumption only modestly.

VII.  Conclusion

The recovery in the current downturn has been disappointing.  GDP has grown since soon after Obama took office, but has grown only slowly, and has been on a path well below that seen in other recoveries.

There are a number of reasons for this.  Household consumption has kept to a low path as households have struggled to repay the over-indebtedness they had accumulated during the housing bubble years.  Residential investment collapsed as well following the bubble, is only now starting to recover, and remains far below the levels seen at similar points in other recoveries.
And government spending has been allowed to fall during Obama’s term.  This had never happened before in the previous downturns.  Indeed, while real government spending rose by 26% at the same point in the economic recovery during the Reagan presidency, it has been reduced by over 1% in this recovery (and reduced by 7% from what it had been a half year after Obama took office).
The reduction in government spending reduced the demand for what the economy could have produced.  In this it was similar to the reduced demand resulting from lower residential investment or lower household consumption expenditure.  All these reductions in demand reduced GDP, reduced the demand for workers, and hence increased unemployment.  But while residential investment and household consumption can only be influenced indirectly and highly imperfectly by government policy, government has direct control over how much it spends.  That is, government can decide whether to build a road or a school building, and doing so will employ workers and will lead to an increase in GDP.  Hence government spending is a direct instrument that can be used to raise growth and employment, should the government so choose.
Sadly, and in stark contrast to the sharp increase in government spending during the Reagan period that spurred the recovery to the 1981 downturn, US politics during the Obama presidency have instead led to a cut-back in government spending, with a resulting drag on growth.  The disappointing consequences are clear.

Obama and Prices: The Markets Expect Inflation to Remain Low

US Treasury Bond Yields, TIPS, and Expected Inflation, Jan 2, 2003, to Aug 8, 2013

Conservative critics of Obama argue his policies will inevitably lead to high inflation.  A previous blog post on this site showed that in fact inflation during the four years of Obama’s first term had been the lowest over any presidential four year term going back a half century.  Low inflation during Obama’s first term cannot be denied.

The conservative critics respond that while inflation may have been low so far, it is inevitable that inflation will soon rise.  The blog post cited above provides links to several examples of what they have been saying.  But this assertion can be examined as well.  In particular, the financial markets (which the conservative critics generally take as reflecting a sound view on such matters, as the investment returns of such investors will depend on getting this right) can be used to see what at least the markets believe inflation will be going forward.

Since 1997 the US Treasury has been issuing bonds of varying maturities whose principal is indexed to the US CPI price index.  These bonds, known as TIPS (for Treasury Inflation-Protected Securities), provide a return which will be the same in real terms regardless of what inflation turns out to be.  The yields on such bonds can be compared to the yields on regular US Treasury bonds of similar maturity.  Such regular US Treasury bonds will pay interest and at the end return the principal in certain dollar amounts, with a value in real terms which will vary depending on what inflation turned out to be.

Inflation is normally positive, so the regular bonds will pay rates which are higher than the rates on TIPS bonds.  But whether the higher rates are worthwhile will depend on how high inflation turns out to be.  To illustrate with some simple numbers, suppose the rate on a 10-year regular US Treasury bond is 3% while the rate on a 10-year TIPS is 1%.  If inflation turns out to be 2%, the bonds will be equally valuable.  But if inflation turns out to be 3%, it would have been better to have invested in the TIPS.  The TIPS will still pay out a 1% real return, while the regular US Treasury will yield a real return of only 0% (a 3% nominal return, but with 3% inflation the real yield will be zero).  Alternatively, suppose inflation turns out to be 1%.  The real return on the regular US Treasury will be 2% (equal to 3% minus 1%), while the TIPS will still yield the contracted 1% real return.  In one believes inflation will be just 1% over this period until maturity, it would have been better to have invested in the regular bonds.

The investors will therefore need to determine what they expect inflation to be.  They will bid up the price of one of the bonds (and bid down the price of the other) if they believe inflation over the time to maturity of the bond, will be higher or lower than the current gap in the yields between the two.  Where the prices of the two bonds settle, and therefore what the gap in yields is between the two, therefore reflects what the financial markets as a whole believe will be the rate of CPI inflation over the period until the bonds mature.  Since real money is riding on this, the investors will take it seriously.

The graph above shows the yields on regular 10-year US Treasury bonds (in blue) and on 10-year TIPS (in green), for the period from January 2, 2003, to August 8, 2013.  The data comes from the official US Treasury web site (where the data presented there goes back to January 2, 2003).  The implied 10-year expected rate of inflation (in red) is then calculated based on the difference between the two yields.

As can be seen in the graph, the yields on the regular 10-year US Treasury bond varied a fair amount over the period, from generally between 4 and 5% during the Bush presidency, falling over time to below 2% for much of 2012, and then rising to about 2 1/2% recently.  The 10-year TIPS yield similarly varied from around 2% during the Bush years, to negative levels for 2012 and the first half of 2013, and rising to a still low but positive 1/2% recently (and most recently just 1/3%).

Despite such fluctuations in the yields of the regular 10-year bond and the 10-year TIPS, the implied expected inflation rate (the difference between the two yields) has been relatively constant, at about 2 1/2% during the Bush years and a similar but slightly lower rate (on average) during the Obama presidency.  The one exceptional period, which should be excluded, would be during the period of economic and financial collapse in the final months of the Bush presidency, after Lehman Brothers went bankrupt and the financial markets were in chaos.  The TIPS yields went up while the regular US Treasury bond yields fell sharply, leading to an implied expectation of inflation of close to zero.  But the figures under such chaotic conditions should not be taken as meaningful.  The chaotic markets then stabilized within a short period of Obama taking office in January 2009, with the rates then returning to more normal levels.

The financial markets, which the conservative critics of Obama normally place a good deal of faith in, therefore do not show any indication that they expect inflation over the next decade to rise.  Rather, they expect inflation of around 2% a year to continue, which is consistent also with the rate of inflation the Fed targets.

Finally, the figures on the bond yields in the graph above also show that the US government has been able to borrow, and continues to be able to borrow, at incredibly low rates, whether in real or nominal terms.   The TIPS yield (the borrowing rate in real terms) was indeed negative in for most of 2012 and the first half of 2013, and is still only 1/2% or less.  Even were it not for the still high unemployment in the country, this is the period when the government should be undertaking investments in both new infrastructure and other assets, and in maintenance of existing assets.  Such investments are worthwhile even if they generate returns of only 1/2% in real terms.  Yet such investments, particularly in maintenance, will generate returns that are orders of magnitude greater than that.

It has been incredibly stupid that the Republican insistence on cutting government spending has blocked us from proceeding with such investments at a time when the borrowing costs to fund them have been so low.

Taxes on Corporate Profits – Low, Falling for Decades, And Now Close to a Voluntary Tax

Corporate Profit Taxes as Share of Corp Profits, 1950-2013Q1

Conservatives bemoan the US corporate profit tax rate, which at 35% for the statutory rate  is the highest among OECD members.  They insist the tax, which they consider to be high, is both unfair and harms US competitiveness.  While they acknowledge that the rates the US corporates actually pay are less, due to legal deductions and other mechanisms, what might not be clear is how low US corporate profit taxes have become.

The US Government Accountability Office (GAO, the audit and investigating agency that  works for the US Congress) released on July 1 a report on what US corporations in fact pay in corporate profit taxes.  Using tax return data (but aggregated to preserve confidentiality), the GAO found that profitable US corporations in 2010 paid federal corporate profit (also called income) taxes at a rate of just 13%, despite the statutory rate of 35%.

If one includes corporations that reported a loss in 2010 (and hence had zero or only little profit taxes due, leaving the numerator the same but whose losses then reduce the denominator in the ratio), the average federal tax rate came to only 17%.  Furthermore, the total tax including not just US federal taxes, but also US state and sometimes local taxes as well as profit taxes paid abroad, came to only 17% in 2010 for the corporations reporting profits, and 22% when one includes the loss-makers.

All these rates are far below the statutory federal corporate profit tax rate of 35%, which has been in place since 1993.  There are state and sometimes local corporate profit taxes on top of this, with rates that vary from zero in certain states (such as Nevada), up to 12% for the top marginal rate (in Iowa).  The state taxes average about 6 1/2%.   Taking account of just federal and state taxes, the corporate profit tax rate on average should be over 41%.

The GAO investigation was carefully done, and has raised again the point that while the US corporate profits tax rate might appear to be high, it bears little relationship to what corporations actually pay.  And the trend over time is decidedly downward.  The graph above uses data from the National Income and Product (GDP) Accounts, produced by the BEA of the US Department of Commerce to show what corporate profit taxes have been as a share of corporate profits since 1950.  While these figures will not be exactly the same as what actual tax return data will show (due to definitional differences in what is included in taxes and especially in how corporate profit is defined, as well as due to timing differences arising from the distinction between when tax obligations are accrued and when they are paid), the trend is clear.  Corporate profit taxes as a share of corporate profits have been falling steadily, from over 50% in 1951 to only 20% recently.  These estimates from the GDP accounts are consistent with the recent GAO figures based on tax return data, where one should note that the BEA estimates will include loss-making firms as well as profitable ones in their averages.

The fall in the actual rate paid to just 20% in recent years also undermines the argument that a high US corporate profits tax rate has undermined the incentive to produce.  Economic performance was better when the profits tax rate paid was much higher than now.  The US corporate profits tax rate averaged 44% in the 1950s and 1960s, yet economic growth was strong then.  As an earlier post on this blog discussed, economic growth performance in the US was substantially better in the 30 years before 1980 than in the 30 years after.

Furthermore, there is little support in the figures that the US corporate profits tax rate at 35% puts the US at a competitive disadvantage vis-a-vis the other OECD members.  While the 35% rate is indeed the highest, the second highest is 34.4% and the third is 33.99% (see the OECD source cited above).  Fifteen of the 33 OECD members covered had rates of 25% or above, and a further 10 had rates of 20 to 24.9%.  More importantly, all of these OECD members, other than the US, imposed a value-added tax on top of their corporate profits tax (and other taxes).  These additional value-added taxes were as high as 27%, and 23 of the 33 OECD members (including essentially all of Europe) had value-added tax rates of 18% or more.  Value-added taxes will be taxes on corporate profits (as well as on labor income), and should not be ignored when one is looking at the overall rate of tax on corporate profits.

The ability to avoid taxes on corporate profits has been receiving increasing attention in recent months.  Historically, much of this avoidance has been achieved through explicit provisions written into the tax code by Congress for certain subsidies or other government expenditures, which the Congress did not want to explicitly provide for or acknowledge in the budget.  Examples include credits for investing in certain locations or for certain purposes (such as R&D), or accelerated depreciation allowances as a mechanism to spur investment.  The objectives might well be worthwhile, but by hiding in the tax code what are in reality subsidies, and then keeping them secret due to the privacy of tax return data, such subsidies are likely to be both inefficient and misguided.  If subsidies are warranted, it would be better to provide them openly and transparently through the budget.

More recently, large corporations have learned how to use international operations as a means of hiding profits from jurisdictions where they would be subject to tax.  Some examples of what US firms have done in the UK to avoid paying taxes there have been recently in the news, and provide good examples of what modern firms can do anywhere, including in the US.

Transfer pricing, while technically illegal, has historically been one mechanism to do this.  This appears to have been one of the ways (among others) that Starbucks was able to run highly profitable coffee shops in the UK, but pay nothing in UK profit taxes.  Starbucks of the UK would “purchase” coffee beans from a Starbucks subsidiary legally based in Switzerland, which would in turn purchase the coffee beans from around the world.  Since commodity trading in Switzerland pays very little tax on the corporate profits generated in such trading, Starbucks could pay the international price for coffee beans through its Swiss subsidiary (even though the beans would never pass physically through Switzerland), and then charge the UK subsidiary a higher price for the beans.  This would increase the costs (and hence reduce the profits) of the UK subsidiary, while generating high profits on coffee bean trading in its Swiss subsidiary, where little or no tax was due on such operations.  And this could be done in essentially any jurisdiction which does not tax corporate profits.

There were other mechanisms as well that Starbucks appears to have used, including intra-company loans from one subsidiary (based in a low tax jurisdiction) to a subsidiary in a jurisdiction (such as the UK) where corporate profit taxes would be due.  This is very similar to transfer pricing on supplies, although here it would be for the supply of capital.

Through these and other mechanisms, Starbucks has been able to avoid, probably legally given the tax code as written, most corporate profit taxes on its UK operations, even though it had consistently reported to analysts on Wall Street that its UK operations were highly profitable.  This became such a public relations disaster that in June the company announced that it would voluntarily pay UK profit taxes of £10 million in 2013 and a second £10 million in 2014.  Starbucks has shown how corporate profit taxes have become in reality voluntary taxes, paid only to avoid image problems.

Starbucks provides a good example of what modern corporates can do, even though it operates just a simple business of selling coffee.  High-tech firms such as Apple are more often in the news since they have generated high profits yet have legally been able to avoid paying taxes on these profits at anything close to the 35% statutory rate.  For example, and as reported in a recent Senate investigation, Apple was able to exploit a difference in how corporations are defined in terms of their tax liability in a country, in order to generate profits in an Irish subsidiary which would not be subject to tax in either Ireland or the US.

More generally, US corporates do not have to pay US corporate income tax on profits generated in overseas operations until these profits are brought back to their US companies.  This is unlike the case for US citizens, who must pay each year income taxes on income generated everywhere in the world, and not just the US.  Because of this provision in the US tax code, Apple and other US corporates have kept accumulated profits legally overseas, so as to avoid paying US profit taxes on them.   The total for large US corporates reached an estimated $1.9 trillion as of the end of 2012, with Apple alone accounting for $102 billion.  Republicans have pushed for a tax amnesty on the repatriation of such funds, as was done once during the presidency of George W. Bush.  But such tax amnesties of course then generate the incentive to hold such profits in untaxed offshore accounts again, in the expectation that an administration in the future will once again grant such an amnesty.

And it has now become straightforward to structure a system of corporate subsidiaries so that almost any company can, if it wishes, make it appear that profits in the US (or indeed any other country, where corporate profit taxes would be due) are close to zero, and instead are high in some low tax or even untaxed jurisdiction such as the Cayman Islands.  Transfer pricing is one such mechanism, although technically illegal as prices between corporate subsidiaries are supposed to be “arms-length” market prices.  But these are effectively impossible to enforce.  How does a tax-audit determine what the price should have been for some specialized input (such as a component going into an iPad), for which no market exists?

But there are other means as well.  High tech firms such as Apple can, for example, transfer ownership of some patent to an Apple subsidiary in the Cayman Islands, and then require the Apple US firm to pay a royalty to the Apple Cayman Islands firm.  Or Starbucks can transfer ownership to the Starbucks brand name similarly to a subsidiary in some low tax or no tax off-shore jurisdiction, and then have the US subsidiaries pay that off-shore subsidiary for the use of that brand name.

The legal “technology” for corporate tax avoidance has therefore come to the point where what is in fact paid in corporate profit taxes can be close to voluntary.  Starbucks in the UK is the most clear case so far.  Governments are concerned, as these mechanisms can now undermine, quite legally, collections on what was at one point an important tax.  The OECD now has a working group looking at possible reforms to address the currently legal ability of modern corporates to avoid taxes through their international operations, with a report scheduled to be released in July.  But it remains to be seen whether politically possible changes in the tax code will be able to ensure such loopholes are closed.