Part Time Workers and the Affordable Care Act: A Proposal to Address the Real Issue

Part Time Workers as Share of Total Employed, Dec 2007 to Dec 2014

A.  Introduction

The Affordable Care Act (ACA, and also often referred to as ObamaCare) has been working well by any objective measure.  There are now more than 10 million additional Americans who have health insurance who could not get affordable health care before; the share of the uninsured in the US population is now a quarter less than what it was before the individual mandate of the Affordable Care Act went into effect; and this has been achieved at premium rates for the new plans that are reasonable and well less than opponents charged they would be.  Health care costs have also stabilized under Obama, both as a share of GDP and in terms of health prices relative to overall prices, in contrast to the relentless increases in both before.  And while some have criticized this, it is good that there are now minimum quality and coverage standards in health insurance plans.  Such standards are good in themselves.  And without such standards, purported health care “plans” which offer next to nothing (due, for example, to extremely high deductibles) and which can then cost next to nothing, would lead to a death spiral for genuine health care plans that cover costs when you are sick and need treatment.

Gains from the ACA are also reflected in the findings of a recently published report from The Commonwealth Fund.  The Commonwealth Fund has been organizing a periodic survey on health care coverage since 2001.  The most recent survey (for 2014) found that for the first time since the question was first asked in 2003, there was a reduction in the number of Americans avoiding (because of cost) health care services that they needed.  And for the first time since the question was first asked in 2005, the number reporting medical bill or debt problems also fell.  Personal financial distress due to medical problems has been reduced, due to greater access to health insurance and due to health insurance plans that now meet minimum standards.

Despite this (but not surprisingly given the position they staked out against the reform), the Republican Congress continues to vote to repeal, or at least weaken, the law.  The most recent vote was aimed at the provision in the Act which complements the individual mandate to purchase health insurance, with an employer mandate requiring firms with 100 full time equivalent employees or more from January 1 of this year (and with 50 or more from January 1, 2016) to offer health insurance to their full time employees or pay a fee.  The proposed Republican bill would change the definition of a full time worker from one who normally works 30 hours or more a week, to one who works 40 hours or more a week.

The supporters of the change charge that the prospect that employers (with 50 or 100 employees or more) will soon be required to offer health insurance to their full time employees has led firms to cut working hours of their employees, to shift them from full time to part time status, and hence avoid the employer mandate of the ACA.  As a Republican congressman from Texas said:  “We have heard story after story from every state in the union that employers are dropping workers’ hours from less than 39 hours a week to perhaps less than 29.”

This accusation is confused on several levels.  This post will first look at whether there is in fact any evidence that workers are being shifted from full time to part time status as a result of the ACA (or indeed for any other reason).  The answer is no, at least at the level of the overall economy.  Second, there has been a good deal of confusion in the discussion on what the issue really is with regard to part time workers, including by prominent congressmen such as Paul Ryan.  Either Ryan does not understand what the employer mandate is, or if he does, then he has deliberately mischaracterized it.

The public discussion has also avoided altogether the real issue.  It is not that firms with 50 workers or more would be required to offer health insurance to their employees (most do already), but that this insurance is only made available to their full time workers.  Part time workers get nothing, no matter what size firm they work at.  The final section of this blog post will discuss a way to resolve this equitably.

B.  What is the Evidence on Whether the ACA Has Increased the Ranks of Part Time Workers?

The opponents of ObamaCare assert that as a result of the employer mandate, firms have been shifting workers from full time to part time status.  E.g., instead of employing one worker for 40 hours, they are choosing to employ two workers for 20 hours each.  If true, the ratio of part time workers to the total employed will rise.

The chart at the top of this post shows this has not been the case.  It is based on data from the Bureau of Labor Statistics, from its Current Population Survey.  This monthly survey of households is used to determine the unemployment rate among other statistics.  The households surveyed are asked whether household members are employed full time or part time (if employed), and if part time, whether this is by choice (because they only want to work part time) or because they want a full time job but cannot find one.  The chart above shows the ratio of workers who are working part time not by choice but for economic reasons, to all workers employed.  Note that the BLS data defines a part time worker as one with fewer than 35 hours of work per week.  While this differs from the 30 hour standard in the ACA, as well as the 40 hour standard in the recently passed Republican legislation, the results in terms of the trends should be similar.  The BLS does not publish data with a different cutoff in terms of hours per week for what is considered part time work.

As in any economic downturn, the ratio rose rapidly in the economic collapse of the last year of the Bush administration.  Regular jobs were disappearing, with some of them shifting to part time status.  Indeed, the absolute number of part time jobs was increasing at the time, even as the total number of jobs was falling, thus leading to two reasons for the ratio to rise, and rise rapidly.

The ratio reached a peak soon after Obama took office, and began to fall about a year later.  Since then it has fallen at a fairly steady pace in terms of the trend.  There were sometimes relatively sharp month to month fluctuations in the data, but this can be on account of statistical noise.  The data comes from a limited sample of households, with only 5 to 6% or so of those employed on part time status (for economic reasons) for most of this period, so the statistical noise in a relative sense (month to month) will be large.  But the downward trend over time is clear, and at a similar downward pace for close to five years now.

What one does not see is any shift in this downward trend linked either to the signing of the Affordable Care Act in March 2010, or to the start of the individual health insurance mandate in January 2014, or to the anticipation of the start of the employer health insurance mandate in January 2015.  Note that since the classification of a worker as a full time or part time worker (and hence the classification of the firm as crossing the 100 or 50 full time worker standard) will be in a period of up to 12 months before the employer mandate goes into effect, one would have seen an impact in 2014 if the 2015 mandate mattered.  There is no indication of this.

The data cover the overall economy.  The figures refer to millions of workers as well as millions of employers.  The US is a large place.  Within such a large place, it will undoubtedly be possible to find particular cases where employers will say that they reduced worker hours to part time status so that they could avoid the health insurance employer mandate.  And one could indeed probably find a long list of firms making such statements.  It would be even easier to find a long list of firms and other entities where working hours were cut, whether or not there was any employer mandate pending.  In a dynamic economy, there will always be a large number of such cases (along with a large number of cases of firms going in the opposite direction, converting part time jobs to full time jobs).

Such anecdotal information, and even a long list of such anecdotes, is not evidence of an issue of substantial scale.  As seen above, there is no evidence of it in the overall numbers.  But one should still recognize that the issue could exist in particular cases.  The question, however, is what is the real issue here, and if there is one, how can it be addressed.

C.  What the Employer Health Insurance Mandate Says

For better or worse, the US health care insurance system is built around health plans normally provided to workers through their place of employment, as part of their overall wage compensation package.  The system began during World War II and has expanded since, supported through substantial tax advantages.  By now, health insurance provision is close to universal among large employers, but substantially less so among small private firms:

Share of Private Firms Offering Health Insurance – 2013
< 10 employees 28.0%
10 to 24 employees 55.3%
25 to 99 employees 77.2%
100 to 999 employees 93.4%
≥ 1000 employees 99.3%
< 50 employees 34.8%
≥ 50 employees 95.7%
All private employees 84.9%
Source:  MEPS, Tables I.A.2 and I.B.2 (2013)

Overall, 84.9% of private sector employees are in firms that offer health insurance as part of their wage packages.  And 96% of firms with more than just 50 employees offer health insurance.

The Affordable Care Act built on this and did not replace it.  Liberals (including myself) would have preferred moving to a system where Medicare would be extended to cover the entire population rather than just those over age 65.  Medicare is an efficient and well managed program, and as an earlier post in this blog discussed, its administrative expenses come to only 2.1% of the benefits paid.  In contrast, administrative costs (including profits) of private health insurance are seven times higher at 14.0% of benefits paid, and an even higher 18.6% of benefits paid in the privately administered Medicare Advantage plans.

But Obama agreed instead to support an approach first proposed by the conservative Heritage Foundation, which was then put forward by Republicans in Congress as their alternative to the health reforms proposed by the Clinton administration (coming out of the task force Hillary Clinton chaired), and which was later adopted in Massachusetts when Mitt Romney was governor.  These plans were built around keeping the existing employer-based provision of health insurance for most of those employed, but to complement this with markets where individuals could purchase health insurance directly if they did not have employer-based coverage, coupled with an individual mandate to buy such health insurance.  The individual mandate is necessary to counter what would otherwise be a resulting death spiral of health insurance plans if everyone is granted access (including those with pre-existing conditions) but only the sick then purchased health insurance (for a description and discussion, see this earlier Econ 101 blog post).

It was not unreasonable to believe that the Republicans would not oppose a plan whose origins lies in their own earlier proposals, but that was not to be.

As noted, the individual mandate is necessary to avoid death spirals in health insurance plans for individuals.  Complementing this, an employer mandate to offer health insurance to their employees is necessary to counter what could otherwise be a “race to the bottom”.  If certain firms did not support such health insurance for their employees, thus reducing the cost to them of their workers, they could undercut competitors who did provide good health insurance support.  It could lead to a race to the bottom.  While not yet widespread in the US, especially for larger firms (see the table above), there has been increasing competitive pressure in the US over the last couple of decades to cut such health insurance support.  An increasing number of employers have done so.

Thus the ACA includes an employer mandate to complement the individual mandate.  However, while the individual mandate went into effect on January 1, 2014, the employer mandate has been twice delayed, and has now (as of January 1, 2015) gone into effect for firms employing 100 of more full time equivalent employees, and will go into effect on January 1, 2016, for firms employing 50 or more full time equivalent employees.  It is this provision that the Republicans in Congress are now trying to subvert.

The charge by Paul Ryan and others has been that medium to small size firms have been cutting the hours of their employees to shift the workers from a full-time classification to a part-time one.  The aim, they say, has been to reduce the number of their full time workers to below 50 so as to avoid the employer mandate.  For example, in a recent opinion piece published in USA Today, Congressman Ryan wrote:  “The law requires employers with more than 50 full-time employees to give them health insurance.  But because the law defines “full time” as 30 hours or more, employers are keeping employees below that threshold to avoid the mandate entirely.”

However, that is not what the law says.  Precisely to avoid such an incentive, the boundaries on the size of a firm subject to the employer mandate is defined in terms of full time equivalent workers (whether 50 or 100).  That is, if a job is split from one full time worker to two half time workers, the number of full time equivalent workers is unchanged.  The two half time workers count as one full time worker for the purposes of the statute.  Cutting back on the number of hours of individual workers to make them part time will not change the status of the firm when the total hours of labor to produce whatever the firm is producing remains unchanged.  And it would be foolish for a firm to produce and sell less when the demand exists for such sales, simply to avoid this mandate.

There is, however, a critically important issue here which Ryan and his colleagues have not discussed.  While splitting jobs of full time workers into multiple part time jobs will not change the status of the firm on whether it is subject to the employer mandate, shifting workers from full time to part time status does affect whether the firm would be required to include health insurance as part of their wage compensation package.  Firms subject to the mandate must offer an affordable health insurance plan available to at least 95% of full time (not full time equivalent) workers, or pay a fee.  The fee (of up to $2,000 per year per worker, less 30 workers per firm) is designed to partially offset (and only very partially offset) the cost of health insurance that they are shifting to others.

But such health insurance typically only is provided to full time workers.  This is true even for giant corporations.  Hence a firm can avoid making health insurance available to its workers by shifting them from full time to part time status.  This has always been the case, and is indeed a problem.

The Affordable Care Act addresses the issue only partially and tangentially.  By including a definition of what constitutes full time work at 30 hours a week or more, the ACA reduces the incentive to shift workers from the traditional 40 hours per week for full time work, to just under 40 hours in order to avoid providing health insurance cover.  A firm would need to cut a normal worker’s hours to below 30 hours per week to avoid providing health insurance, and is unlikely to do that for its regular work force.  But by moving the dividing line up to 40 hours per week, as the Republican legislation passed on January 8 would do, one opens up a loophole for firms to reduce worker hours from 40 to say 39 per week (or 39 1/2 or even 39.99 I would suppose).  Firms would be able easily to avoid offering health insurance to what are in reality their regular, full time, workers; use this to undercut competitors who do offer such insurance; and thus spark a race to the bottom on health insurance coverage in those industries.

D.  Addressing the Problem of Health Insurance for Part Time Workers

As noted above, the ACA does not do much to address the problem of part time workers receiving nothing from their employers for the health insurance everyone needs.  Setting the floor at 30 hours per week helps by ensuring workers close to the traditional 40 hour workweek will receive an employer contribution to their health insurance, and avoids the incentive to shift workers from 40 hours per week to just a bit below.  But part time workers of less than 30 hours per week will still normally receive nothing from their employer to help cover their health insurance.  And it creates an incentive for employers to structure positions as two workers at 20 hours per week, say, than one at 40.  While whether or not the firm was subject to the employer mandate would not be affected (since it is expressed in terms of full time equivalent workers), whether or not the firms would need to provide anything in terms of health insurance would be affected.

But there is a way to address this, now that the individual health insurance marketplaces are operational under the ACA.  All firms could be required to contribute an amount for their part time workers proportional to the hours of such part time work to what full time work would be.  That is, if two workers are each working half time, the firm would contribute an amount of 50% (for each) of the cost of the employer contribution to the health insurance for one full time worker.  The total cost would be the same whether the firm employed one full time or two half time workers.  There would also then not be an incentive to split jobs from full time workers to multiple part time workers.

The employer contribution to the part time worker’s health insurance costs would then be paid, along with taxes such as for Social Security or Medicare, to the government in the name of the specific part time worker.  These funds would then be used as a partial pay down of the costs of that worker purchasing health insurance on the individual health insurance market exchanges set up under the ACA.  And while other splits could be considered, I would recommend that those funds would be split half and half between what the worker would need to pay on the exchange for his or her health plan, and what the government subsidy would provide.

A simple numerical example may help clarify this.  Using made up numbers, suppose the full monthly cost of a standard (Silver level) health insurance plan on the individual exchange where the worker resides is $400.  Assume also that at the current income level of this (part time) worker, the government subsidy for such insurance would be $200 per month, while the worker would pay $200 per month.  Now assume that firms would be required to pay proportional shares of what they provide to full time workers for their health insurance, and that this would come to $100 per month for this part time worker.  This would be split half and half between what the government subsidy would be and what the worker would pay, so under the new approach the government would provide $150, the worker would pay $150, and the funds coming from the firm would cover $100, summing to the $400 total cost.

A few specifics to note:  Many part time workers hold down multiple jobs.  They would receive for their “account” the total proportional amounts from all of their employers.  Many part time workers are also part of married couples.  There could be a household account into which all the sums were paid (for each family member), which could be used to purchase a family health plan on the exchanges.  In the event that the family was not purchasing insurance through the exchange (perhaps, for example, because the spouse worked at a firm providing family coverage), the amount paid by the firm for the part time worker would be returned to the firm (or canceled from the start).

And if the total amounts paid in from the full set of employers for that individual (or family) led to the government subsidy falling all the way to zero, any excess would be allocated to what the individual would pay for the insurance.  This could be common in cases where the family income of the part time worker was close to, or above, the income limit on which government subsidies are provided.

It is only with the advent of the individual health insurance exchanges that this method for covering part time workers became possible.  Previously, firms were not in a position to purchase half of an insurance policy for a half time worker.  But now they can contribute an amount equal to half the cost, with this then used to help purchase coverage on the individual marketplace exchanges.

Note also that with this reform, it would matter less whether full time work was defined as 30 hours per week or 40 hours per week or whatever.  I would recommend keeping the 30 hour per week boundary as it would be a factor in determining what the employer contribution would be.  But it would not be as critical as now, where the boundary determines whether 100% of the employer share of the health insurance cost is paid or 0% is paid.  There would be a smooth transition (a worker of 39 hours when 40 hours is defined as the standard would still receive 39/40 of the payment, and not zero), without a drop straight to zero.

There would also be no reason to limit this extension of the employer mandate only to firms with 50 (or 100) or more full time equivalent workers.  All firms should make such a contribution to covering the cost of their workers’ health insurance needs, just as they all make a contribution to Social Security and Medicare taxes.  Indeed firms of whatever size (although this will soon apply only to firms with less than 50 full time equivalent workers) that do not have any health insurance plan for their staff should participate.  The amounts paid could be set as a proportion to the cost of the medium Silver level plan available on the individual health insurance exchanges in their area.

Undoubtedly, there will be assertions by the Republicans that requiring such a contribution to health insurance costs for their part time workers will lead to an end to such jobs.  This would be similar to the arguments they have made that raising the minimum wage will lead to higher unemployment of lower paid workers, and arguments that were made earlier that paying Social Security taxes would lead to higher unemployment.  But as was discussed in an earlier blog post, there is no evidence that increases in the minimum wage in the magnitudes that have been discussed have led to such higher unemployment.  Ensuring firms contribute proportionally to the health insurance costs of their part time workers would not either.

The Strong Recovery in Employment Under Obama

Unemployment Rates - Obama vs Reagan

A.  The December Jobs Report

The Bureau of Labor Statistics released on Friday its regular monthly report on employment.  Job growth was once again strong.  Total jobs (nonfarm payroll employment, to be precise) rose by a solid 252,000 in December, and the unemployment rate came down to 5.6%.  Total jobs rose by an even higher (and upwardly revised) 353,000 in November and by an also upwardly revised 243,000 in October (the two most recent monthly figures are always preliminary and subject to revision).  These are all good numbers.  The 353,000 figure for job gains in November was the highest monthly figure in over nine years.

The overall job gain in 2014 came to 2.95 million.  This was the highest annual total since 1999.  Private sector jobs rose by 2.86 million in 2014.  This was the highest annual gain in private jobs since 1997.  Government jobs (federal, state, and local) also grew, although only by 91,000 and equal to just 3% of the overall growth in jobs of 2.95 million.  But at least it was positive and stopped being the drag on growth it had been before through repeated cuts.  Government jobs had been cut each and every year since 2009, reducing American jobs by 702,000 between 2009 and 2013.

B.  Obama’s Performance on Unemployment, Compared to Reagan’s

While the pace of improvement has accelerated in the past year, the Obama record on jobs has in fact been a good deal better for some time than he has been given credit for.  Critics said that Obama’s policies, both as a consequence of the passage of the Obamacare health reforms and from his use of government regulatory powers, would (they asserted) constrain job growth and keep unemployment high.  These critics look to the Reagan presidency as a model, with the belief that there was a rapid fall in unemployment following his tax cuts, attacks on unions, and aggressive deregulatory actions.  This adulation continues.  A recent example was a column by Stephen Moore (Chief Economist of the Heritage Foundation) published in the Washington Post just two weeks ago (and which a number of commentators, including Paul Krugman, noted was full of errors).

But how do the Obama and Reagan records in fact compare?  The graph at the top of this post shows the path the rate of unemployment has taken during Obama’s presidency, and for the same period during Reagan’s presidency.  Both curves start from their respective inaugurations.

Unemployment under Reagan was high when he took office (at 7.5%), although on a downward trend.  But it then rose quickly (peaking at 10.8%) followed by a fall at a similar pace, before leveling off at a still high 7 to 7 1/2%.  It then fell only slowly for the next two and a half years, by a total of just 0.6% points.  The recovery in terms of the unemployment rate lacked strong staying power.

The pattern was different under Obama.  While unemployment was also high when he took office (7.8%), it was rising rapidly as the economy was losing 800,000 jobs a month.  It rose to a peak of 10.0% nine months later, before starting a fall that has continued to today.  The pace of the reduction was relatively steady over the years, but accelerated in 2014.  Over the last two and a half years, the unemployment rate has been reduced by 2.6% points, far better than the 0.6% reduction for the comparable period under Reagan.

As a result, the unemployment rate is now 5.6% under Obama, versus 6.6% at the same point in Reagan’s tenure.  By this measure, performance has been better under Obama than it was under Reagan.  The 5.6% rate under Obama can also be compared to Mitt Romney’s statement in 2012, during his presidential campaign, that adoption of his policies would bring the unemployment rate down to 6% by January 2017.  Romney viewed this as an ambitious goal, but achievable if one would follow the policies he advocated.  It was achieved under Obama already by September 2014.  One did not hear, however, any words of congratulations from Romney or others in the Republican Party to mark that success.

Of possibly more interest in the debate about the response to the respective policy regimes of Obama vs. Reagan, was the flattening out of unemployment under Reagan at the still high level of 7.5% or close to it in mid-1984.  If his “supply-side” policies were going to be effective in bringing down unemployment, this was the period when they should have been working.  The tax cuts had been passed, and the regulatory and other policies of the Reagan administration were being implemented and enforced.  But unemployment was trending down only slowly.  In contrast, unemployment was falling rapidly for the same period in the Obama presidency, with the pace of reduction indeed accelerating in 2014.  There is absolutely no evidence that Obamacare, actions to protect the consumer or the environment, the application of government regulations under Obama, or even “policy uncertainty” (a new criticism of Obama that was given prominence during the 2012 campaign), have acted to slow job growth.

C.  A Few More Points

1)  Why did unemployment rise rapidly from mid-1981 to late-1982 (to 10.8%), and then fall at almost the same rapid rate from that peak to mid-1984?  This had less to do with the policies of Reagan than of those of Paul Volcker, then Chairman of the Federal Reserve Board (and a Jimmy Carter appointee).  Volcker and the Fed raised interest rates sharply to bring down inflation, with the federal funds rate (the interest rate at which banks lend funds on deposit at the Fed to each other; it is the main policy target of the Fed) reaching over 19% at its peak.  Inflation came down, the Fed then reduced interest rates, and the economic downturn that the Fed policy had induced was then reversed.  Unemployment thus rose fast, and then fell fast.

2)  Has the fall in the unemployment rate under Obama been more a reflection of people dropping out of the labor force than a recovery in jobs?  No.  A previous post on this blog looked at this issue, and found that labor force participation rates have been following their long term trends.  There is no evidence that labor force participation rates have made a sudden shift in recent years.

3)  Why has the pace of improvement in the unemployment rate accelerated in 2014?  As earlier posts on this blog have noted,  Obama is the only president in recent history where government spending has been cut in a downturn.  The resulting fiscal drag pulled back the economy from the growth it would have achieved had government spending, and its resulting demand for goods and services and hence jobs to produce those goods and services, not been cut.

But this finally turned around in 2014.  Congress finally agreed to a budget deal with Obama, and state and local governments saw spending stabilize and then start to rise as the recovery got underway and boosted tax revenues.  The result is shown in this chart, copied from an earlier post whose focus was on austerity policies in Europe:

Govt Expenditures, Real Terms - Eurozone and US, 2006Q1 to 2014 Q2 or Q3

Total US government expenditures (federal, state, and local; in real terms; and for all purposes, including both direct purchases of goods and services and for transfers to households such as for Social Security and Medicare), turned around in the first quarter of 2014 and began to rise.  Government spending had previously been falling from mid-2010.  With that turnaround in government spending, GDP rose by 4.6% (at an annualized rate) in the second quarter of 2014 and by 5.0% in the third quarter.  Much more was going on, of course, and one cannot attribute all moves in GDP growth to what has happened to government spending.  But the turnaround in government spending meant that this component of GDP stopped acting as the drag on growth that it had been before.

Jobs then grew in 2014 at the most rapid rate since 1999, and unemployment fell.  The unemployment rate of 5.6% is the lowest since 2008, the year the economy entered into the economic collapse that marked the end of the Bush administration.

D.  How Much Further Does Unemployment Need to Fall to Reach Full Employment?

The 5.6% rate is not yet full employment.  While it might appear to some to be a contradiction in terms, there will always be some unemployment in an economy, even at “full employment”.  There will be frictions as workers enter and leave jobs, mismatches in skills and in geographic location, and so on.  But the 5.6% rate is still well above this.

Historically, the US economy was often able to achieve far lower rates of unemployment and not see excessive upward pressure on wages and prices.  The unemployment rate was at 4.4% in 2006/07, at 3.9% in 2000, and at 4.0% or below continuously from late 1965 through to early 1970 (and reached 3.5% or below for a full year from mid-1968 to mid-1969).  It even dipped to a post-war low of 2.5% in 1953, although few would say that the conditions then would apply to now.

Based on such historical measures, the unemployment rate could still be reduced substantially from where it is now before the labor market would be so tight as to cause problems.  Economists debate what that rate might be at any given time, but personally I would say that a reasonable target would be no higher than 4 1/2%, and perhaps as low as 4%.

But rather than try to predict what the full employment rate of unemployment might be, one can follow a more operational approach of continuing to push down the rate of unemployment until one sees whether upward wage and price pressures have developed and become excessive.  That is how the Fed operates and determines what policy stance to take on interest rates.  And there is absolutely no sign whatsoever that there is such upward wage or price pressure currently, with the unemployment rate of 5.6%.

As a number of the news reports on the December BLS employment report noted, while unemployment has come down, estimated hourly earnings in December also fell by 5 cents from the previous month (to $24.57 for all private non-farm jobs, from $24.62 the previous month).  Such a one-month change is not really significant, and could be due to statistical fluctuations (as the data comes from a survey of business establishments).  But what is significant is that average hourly earnings in recent years have only kept pace with low inflation of less than 2% a year.  In real terms, wages today are almost exactly the same as they were in late 2008.  This has been the case even though labor productivity is about 10% higher now than in late 2008 (this figure is an estimate, as the GDP figures for the fourth quarter of 2014 have not yet been reported).  In a properly functioning labor market, real wage growth will be similar to labor productivity growth.  But high unemployment since the 2008 downturn has weakened labor’s bargaining position, leaving real wages flat.

Government policy, including actions by the Fed, should be to keep the expansion going at as fast a pace as possible until unemployment has fallen so low that one sees upward pressures on wages and hence prices.  As I noted above, I would not expect to see that until the unemployment rate falls below 4 1/2%, and quite possibly below 4%.

The Cost of Health Care Has Stabilized Under Obama

Total National Health Expenditures as Share of GDP, 1980-2013

A.  Introduction

The Centers for Medicare and Medicaid Services (CMS) released in early December its regular annual estimate of overall health care expenditures in the US.  Their highly detailed tables start in 1960 and now go through 2013, and they provide the most reliable and complete regular figures on health care spending in the US.  While a number of news outlets noted that national health care expenditures had once again remained stable at 17.4% of GDP under Obama (for the fifth straight year now), there is much more that one can derive from these numbers that is of interest to anyone concerned with US health care expenditures.

B.  National Health Care Expenditures as a Share of GDP

The stability of total national health care expenditures at 17.4% of GDP under Obama is indeed significant.  But it is not unprecedented:  Health care expenditures were also stable as a share of GDP for an extended period during the Clinton administration.  But the general path has been strongly upward over recent decades, with the share now close to double what it was in 1980.  Large increases during the Reagan/Bush I and Bush II presidential terms were not offset by the stability during the Clinton and Obama years.  While I have not examined in detail the primary reasons for this difference, I would suspect that a factor has been the greater willingness during Democratic administrations to use government initiatives to hold down health costs.

But while the share of health expenditures in GDP in current prices has almost doubled over this period, the share expressed in terms of constant prices has been flat.  That line is also shown in the chart above, in red.  While there is no published estimate of a price deflator specifically for overall national health expenditures, it is reasonable to use the price deflator in the GDP accounts for personal consumption of health care.  The personal consumption figure accounts for about two-thirds of national health care expenditures, where the remainder will be for such items as investment in hospitals and equipment, for direct government expenditures on health care such as for doctors in the military and in the Veterans Administration, and for research.

Using this price deflator, the share of health expenditures in GDP in real terms in fact declined some over 1980 to 2000, rose by an equal amount between 2000 and 2009, and since then has been flat, to end in 2013 at the same share as it was in 1980 (8.9% of GDP in terms of the prices of 1980).  This is pretty remarkable.  Despite an aging population over this period, where older people require much more health care services than younger ones do, US spending on health care as a share of GDP would have been no higher in 2013 than it was in 1980 if the price of health care relative to overall prices (the GDP deflator) had not changed.

Note that this is not a result of the prices of 1980 as being something special.  The same result would have been found using the prices of any year.  And while not shown in the diagram above, the constancy of the share of health expenditures in GDP in real terms held back to the mid-1970s.  The share rose from the mid-1960s to the mid-1970s, in part due to the introduction of Medicare (the Medicare Act was passed during the Johnson administration in 1965, and the program started in 1966).  The increase in share over that period was by about a quarter (from a bit over 7% to a bit less than 9% of GDP, all in terms of 1980 prices).  It has since been relatively constant.

C.  Relative Prices Matter

The GDP share could only rise in current prices when it was flat in constant prices because the price of health care rose relative to the general price deflator for GDP.  This is just arithmetic.  It is therefore of interest to look more closely at what has happened to the relative price of health care.

For the period since 1980, health care prices have consistently out-paced the rise of overall prices until the last few years:

Change in Relative Price of Health Care vs. GDP, 1980-2013

 

The price index for GDP is a weighted average of the prices of all goods and services produced by the economy.  That is, and speaking loosely, a GDP price index rising by say 2% implies that about half (in weighted terms) of all prices rose by more than 2% while about half rose by less than 2% (including some that could have fallen).

What is unusual for the health care price index is that it has risen consistently faster than the overall GDP price index, until recent years.  The increase was particularly rapid during the Reagan / Bush I years, with the health care price index outpacing the GDP price index by 4.1% per year on average over this period.  For the more technically minded, the GDP deflator rose at an annual average rate of 3.9% over this period, while the health care price index rose at an annual average rate of 8.2%, so the relative price rose at the rate of 1.082/1.039, which equals 1.041, or 4.1% a year.

A 4.1% relative price growth compounded over 12 years (1980 to 1992) is huge:  At that rate, health care prices rose by 62% more than overall prices over that 12 year period.  And that is the immediate cause of health care rising as a share of GDP from 8.9% to over 13% in current prices over the period, despite a slowly falling share in real terms.  Real health care consumption relative to GDP fell, but total health care expenditures still rose relative to GDP in current dollar terms due to the higher relative prices for health care.

The relative price of health care relative to GDP then continued to rise, but at a much slower pace, during the Clinton years.  It then bounced back up some during the Bush II years (other than in 2005 and 2006, when the GDP deflator rose in the peak years of the housing bubble and then matched the increases in the price deflator for health care in those two years).

Under Obama, the relative price of health care came back down, and indeed was significantly negative in 2011 for the first time since before 1980.  This was then followed by two further years of zero or negative growth.  There have not been three consecutive years zero or negative growth in the relative price of health care in the US since 1946 to 1948, two-thirds of a century ago.

The Obamacare reforms account for at least some of this.  The Affordable Care Act (Obamacare) was passed in early 2010, and while the insurance coverage reforms (making health care insurance coverage available for all Americans) only went into effect in 2014, other health care reforms went into effect immediately.  These included a wide range of individually modest, but cumulatively significant, measures to bring down costs.  For example, the Medicare system for compensating hospitals now is set up to provide a financial incentive for good rather than poor quality care.  Under earlier systems, hospitals were paid more when the patient received poor quality care and got an avoidable infection, for example.  Such measures improved efficiency and brought down costs.

D.  Even At a Constant Share of GDP in Real Terms, Per Capita Consumption of Health Care Can Still Rise 

The relative price of health care has stabilized for three years now under Obama, while the share of health care expenditures in GDP, whether in real or nominal terms, has stabilized for five years.  But has this been achieved at the cost of reducing the availability and use of health care?  No:

Growth of Real Per Capita Personal Consumption of Health Care, and of Real GDP, 2001-2013

This diagram plots what has happened since 2001 to real per capita national health expenditures (from the same figures as used above from the CMS, but now converted into real per capita terms), real per capita personal consumption of health care services (from the GDP accounts), and real per capita GDP.  The figures are all scaled to equal 100 in 2008.  The national health expenditure and personal consumption of health care lines track each other fairly closely.  One could have used either.

As the graph shows, real per capita expenditures on (or use of) health care services have increased each year over this period.  There was still an increase, although at a slower pace, in the peak years of the economic downturn in 2009 and 2010.  And the increases continued, at a strong pace, in 2011 to 2013, when GDP was recovering as well.

When health expenditures stabilized as a share of GDP under Obama, some analysts at first speculated that this was due to lower consumption of health care services during the economic downturn.  Unemployment was high and many had less access to health insurance.  But use of health care services did not fall during the downturn.  And it then came back strongly in 2011 to 2013.  The stable share of GDP has been due to stable prices for health care since 2011, with real per capita health care expenditures then rising at a similar rate as rising real per capita GDP.

E.  Why Isn’t the Figure for National Health Expenditures Equal to 18% of GDP? 

An earlier post in this blog in the series on health reform stated that the US has been spending close to 18% of GDP on health care.  This was 50% more than the second highest spending OECD country (the Netherlands) and close to double the average spent of all OECD countries.  The figures were for 2011 and came from the then current OECD data for the US and other OECD countries (close to, but not quite the same as, the national health expenditure totals from the CMS for the US).  Why are the figures for the US now at 17.4% of GDP in 2011, as well as since?

The US health expenditure numbers have in fact not changed.  They are still expected to total $3 trillion in 2014.  The reason for the difference (aside from round-off:  they were a bit below 18% in the earlier numbers) is that the estimate of the denominator in the health expenditures to GDP ratio has changed.  In the summer of 2013, the BEA revised its methodology for estimating GDP, as it periodically does.  While there were several changes, the one with the largest impact was to revise the treatment of research and development expenditures.  The BEA had before treated such expenditures as what economists call an intermediate product (a good which is immediately used up as goods are produced, much like coking coal is used up in the production of steel).  They decided it was more appropriate to treat them as an investment product, which will last for several years (depreciating over time).  This was purely a methodology change.  But the effect was to revise estimated GDP up by about 3 to 3 1/2% in recent years.  This was not just applied to the GDP figures of recent years, but rather to the full GDP series going all the way back to 1929.  Hence the year to year growth rates were largely unaffected.

But a denominator which is now larger will lead to a health expenditure share in GDP which is lower.  By simple arithmetic, a share of 17.9% of GDP will fall to 17.4% of GDP if GDP is estimated to be 3% higher than before.

F.  Conclusion

Health care costs stabilized during Obama’s tenure, with health care costs as a share of GDP now flat (in both constant and in current prices) in contrast to the big increases (in current prices) before.  This has not come at the expense of falling availability or use of health care services.  They have continued to grow throughout his presidency, and especially since 2010.

Looking forward, 2014 may be different.  The Obamacare insurance reforms came into effect in 2014, and have reduced the ranks of the uninsured by more than 10 million Americans.  The share of the population without any health insurance fell by over 30%.  The newly insured are likely to make greater use of regular health care services in 2014, especially by those who previously had conditions which had been left untreated due to an inability to pay before.  However, this may be offset by fewer emergency room admissions by those who previously had no other option, where emergency room care is an especially expensive way to deliver health care services.

It is not clear what the net effect will be.  Preliminary quarterly GDP data (for the first three-quarters of 2014) do not show a rise in the share for personal consumption of health care (there was a growth in real terms similar to the growth in real GDP).  But these numbers are still early and preliminary.  And the full national health expenditure numbers for 2014 will not be out until next December.

But so far, health expenditures as a share of GDP have stabilized under Obama, and the preliminary indication is that this is continuing in 2014.  This is a major achievement.  But they have stabilized at what is still a very high share of GDP, far higher than what is spent on health in other OECD countries.  Much more aggressive and fundamental reform will be necessary to bring the share down to the far lower levels of what other countries spend, and yet obtain  health outcome results that are similar to or better than the outcomes in the US.

How Much Was Bank Regulation Weakened in the New Budget Bill? And What Can Be Done Now?

A.  Introduction

In a rare, late-night and weekend, session, the US Senate on Saturday night passed a $1.1 trillion government funding bill to keep the government running through to the end of fiscal year 2015 (i.e. until September 30, 2015).  The House had passed the bill on Thursday, and it has now gone to Obama for his expected signature.  Had it not been passed, the government would once have been forced to shut down due to lack of budget authority.  It was a “must-pass” bill, and as such, was a convenient vehicle for a number of provisions which stood little chance to pass on their own, but which could only be blocked by opponents now at the cost of forcing a government shutdown.  The specific provisions included were worked out in a series of deals and compromises between the leadership of the Republican-controlled House and the now Democrat-controlled (but soon to be Republican-controlled) Senate.

One such provision was an amendment to the Dodd-Frank Wall Street reform bill, originally passed in July 2010, which enacted a series of measures to strengthen the regulatory framework for our financial sector.  The failure of this framework had led to the 2008 economic and financial collapse in the last year of the Bush administration.  The amendment in the new budget bill addressed just one, and some would say relatively minor, provision in Dodd-Frank.  But its inclusion drew heavy criticism from liberal Democrats, led by Senator Elizabeth Warren of Massachusetts.  Senator Warren argued that the amendments to Section 716 of Dodd-Frank would “would let derivatives traders on Wall Street gamble with taxpayer money and get bailed out by the government when their risky bets threaten to blow up our financial system.”  The amendments were reportedly first drafted by lobbyists for Citibank, and would benefit primarily a very small group of large Wall Street banks.

The amendments do reflect a backwards step from the tighter controls on risk that Dodd-Frank had provided for.  In my view, it would have been better to have kept the original provisions on the issue in Dodd-Frank.  But with a positive response now by the regulators to the reality of this new provision, the impact could be negated.  This blog post will discuss what was passed, what could be done now by bank regulators to address the change, and the politics of it all.

B.  The Amendment to Dodd-Frank, and the Economics of Derivatives

The amendment to Dodd-Frank addresses one specific provision in a very large and comprehensive bill.  An important link in the 2008 financial collapse was the risk major banks had carried on their books from certain financial derivative instruments.  “Derivatives” are financial instruments that derive their price or value from the price or value of some other product.  For example, oil derivatives derive their value from the price of oil (perhaps the price of oil at some future date), foreign exchange derivatives are linked to foreign exchange rates, credit default swaps are linked to whether there is a default on some bond or mortgage or other financial instrument, and so on.

Derivatives can be quite complicated and their pricing can be volatile.  And they can lead to greater, or to reduced, financial risk to those who hold them, depending on their particular situation.  For example, airlines must buy fuel to fly their planes, and hence they will face oil price risk.  They can hedge this risk (i.e. face reduced risk) by buying an oil derivative that locks in some fixed price for oil for some point in the future.  An oil producer similarly faces an oil price risk, but a bad risk for it is the opposite of what the airline faces:  The oil producer gains when the price of oil goes up and loses when it goes down.  Hence both the airline and the oil producer can reduce the adverse risk each faces by entering into a contract that locks in some future price of oil, and derivative instruments are one way to do this.  Banks will often stand in the middle of such trades, as the buyer and the seller of such derivative instruments to the airlines and the oil producers (in this example), and of course to many others.

Derivatives played an important role in the 2008 collapse.  As the housing bubble burst and home prices came down, it became clear that the assumptions used for the pricing of credit default swaps on home mortgages (derivatives which would pay to the holder some amount if the underlying mortgages went into default) had been badly wrong.  Credit default swaps had been priced on the assumption that some mortgages here and there around the US might go into default, but in a basically random and uncorrelated manner.  That had been the case historically in the US, for at least most of the time in the last few decades (it had not always been true).  But this ignored that a bubble could develop and then pop, with many mortgages then going into default together.  And that is what happened.

Dodd-Frank in no way prohibits such derivative instruments.  They can serve a useful and indeed important purpose.  Nor did Dodd-Frank say that bank holding companies could no longer operate in such markets.  But what Section 716 of Dodd-Frank did say was that banks that took FDIC-insured deposits and which had access to certain credit windows at the Federal Reserve Board, would not be allowed, in those specific corporate entities, also to trade in a specifically defined set of derivatives.  That list included, most notably, credit default swaps that were not traded through an open market exchange, as well as equity derivatives (such as on IBM and other publicly traded companies) and commodity derivatives (such as for oil, or copper, or wheat).  The bank holding companies could still set up separate corporate entities to trade in such derivatives.  Thus while Citigroup, for example, could set up a corporate entity owned by it to trade in such derivatives, Citibank (also owned by Citigroup), with its FDIC-insured deposits and with its access to the Fed, would not be allowed to trade in such derivatives.  That will now change.

It is also worth highlighting that under Dodd-Frank, banks with FDIC-insured deposits could still directly trade in such derivatives as interest rate swaps, foreign exchange derivatives, and credit default swaps that were cleared through an organized public exchange.  That had always been so, and will remain so.  The banks could also always hedge their own financial positions.  But they will now be allowed to trade directly (and not simply via an associated company under the same holding company) also in the narrow list of derivative instruments described above.

It is arguable that this is not a big change.  All it does is allow bank holding companies to keep their trading in such derivative instruments in the banks (with FDIC-insured deposits) that they own, rather than in separately capitalized entities that they also own.  The then Chairman of the Federal Reserve Ben Bernanke noted in testimony in front of a House panel in 2013 that “It’s not evident why that makes the company as a whole safer.”

So why do banks (or at least certain banks) want this?  Banks with FDIC-insured deposits and who also have access to certain credit windows at the Fed, are seen in the market as enjoying a degree of support from the government, that other financial entities do not enjoy.  The very largest of these banks may be viewed as “too-big-to-fail”, since the collapse of one or more of them in a financial crisis would in turn lead to a financial cataclysm for the country.  Thus depositors and other lenders are willing to place their money with such institutions at a lower rate of interest than they would demand in other financial institutions.

Thus Senator Warren and others charge that the amendments to Dodd-Frank “would let derivatives traders on Wall Street gamble with taxpayer money and get bailed out by the government when their risky bets threaten to blow up our financial system”, as quoted above.  To be more precise (and less eloquent), any bank with FDIC-insured deposits will make investments with those deposits, those investments will have varying degrees of risk, and if there is a threat that they will fail, the government may decide that it is better for the country to extend a financial lifeline to such banks (as they did in 2008) rather than let them fail.  The amendments to Dodd-Frank will allow these banks to invest in a broader set of derivative contracts directly (rather than only at the holding company level) than they could have before.  Thus they could end up investing directly in a riskier set of assets than they could have before without this amendment, and all else being equal, there could then be a higher risk that they will fail.

Finally, it should be noted that the amendments to Dodd-Frank will benefit largely only four very large banks.  The most recent quarterly report from the Office of the Comptroller of the Currency indicates that just four big banks (Citibank, JP Morgan Chase, Goldman Sachs, and Bank of America) account for 93% of derivative contract exposure among banks (as of June 30, 2014).  While this figure includes all types of derivatives, and not just those on the list that is at issue here, it is clear that trading in such instruments is highly concentrated.

C.  What Can Be Done Now? 

Over the objections of Senator Warren and others, the amendments to Section 716 of Dodd-Frank have been passed as part of the budget bill.  Banks, and in practice a limited number of very large banks, will now be able to take on a riskier set of assets on their balance sheets.  But Dodd-Frank, and indeed previous bank regulation, has established a bank regulatory and supervision regime that requires that banks hold capital sufficient, under reasonable estimates of the risks they face, to keep them out of insolvency and an inability then to repay their depositors.  The bank regulatory and supervision framework was clearly inadequate before, as the 2008 collapse showed.  Dodd-Frank has strengthened it considerably.  The specific rules are now being worked out, and like all such rules will evolve over time as experience dictates.

The amendments to Section 716 of Dodd-Frank will now change the set of risks the banks will possibly face.  I would suggest that now would be a good time for current Fed Chair Janet Yellen, or one of the other senior bank regulators heading up the process or even President Obama himself, to make a statement that they will of course follow the dictate of the law (as spelled out in Dodd-Frank, as it still stands) to take into account these possible new risks as they work out the capital adequacy ratios for the banks that will be required.

Specifically, the statement should make clear that the capital ratios required of banks that trade in these newly allowed instruments will now have to be set at some higher level than would previously have been required, due to the higher risks of such assets now in their portfolio.  It could and should be made clear that the law requires this:  The regulators are required to determine what the capital ratios must be on the basis of the risks being held by the banks in their portfolios.  How much higher the capital ratios will need to be will depend on the riskiness of these new assets compared to what the banks previously invested in, and how significant such new assets will be in their portfolios.  It is quite possible that faced with such higher capital requirements, the banks that had pushed for this new latitude will decide that it would be wiser not to enter into those new markets after all.  They may well come to regret that they pushed so strongly for these amendments to Dodd-Frank.

This is perhaps not the best solution.  The prohibition on direct trading in the proscribed list of certain financial derivative instruments is cleaner and clearer.  Most importantly, while current bank regulators may use their authority to ensure banks hold sufficient capital to reflect the greater risk in their portfolio, there is the danger that regulators appointed by some future president may not exercise that authority as wisely or as carefully.  This was indeed the fundamental underlying problem leading up to the 2008 collapse.  The Bush administration was famously anti-regulation, and Bush appointed officials who were often opposed to the regulations they were in office to enforce.  In at least some cases, the officials appointed who were not even competent to carry out their enforcement obligations.  For example, Bush famously appointed former Congressman Christopher Cox as head of the SEC.  The SEC at the time had the obligation to regulate investment banks such as Lehman Brothers, Goldman Sachs, and Morgan Stanley.  As Lehman Brothers collapsed, with worries that Goldman Sachs, Morgan Stanley, and others would soon be next, Christopher Cox was at a loss on what to do, and was largely by-passed.  Dodd-Frank changed regulatory responsibility (the Fed and other financial regulators are now clearly responsible, where Goldman Sachs and Morgan Stanley had already been “encouraged” to become formal banks and as such subject to Fed oversight), but there is the risk that some future president will choose, like Bush, to put in place figures who either do not believe in, or are not capable of, serious financial supervision and oversight.

In addition, financial crises are always a surprise.  They occur for some unexpected reason.  If they were expected, actions could be taken to address the causes, and they would not happen and hence not be observed.  But surprises happen.  Hence Dodd-Frank, and indeed all financial regulation, includes an overlapping and mutually reinforcing set of measures to try to ensure crises will not occur and that banks will not become insolvent should they occur.  One does not know beforehand which of the regulatory measures might be the critical one for some future and unforeseen set of circumstances leading to a crisis.  it is therefore wise to include what others might call redundancies.  The amendment to Dodd-Frank will remove one of the possibly redundant measures to ensure bank safety.  The remaining measures (e.g. the capital adequacy requirements) may well suffice to address the safety issue, but in cases like this, redundancy is better.

D.  The Politics of It All

While the economics may suggest that the change resulting from the amendment to Dodd-Frank need not be catastrophic if regulators respond wisely, and hence that the amendment is not such a big deal, the politics might be different.

The biggest concern is that many see this as possibly the opening round of a series of amendments to Dodd-Frank and other laws identified with the Obama administration, that a Republican Congress and now Senate will push through on “must-pass” legislation such as budget bills.  Particularly if this had slipped through quietly, with little public attention until after the bill had been passed, the bankers and their Republican representatives could have seen this as a model of how to pass changes to legislation that would not otherwise have gone through.  While the model is certainly not a new one, its affirmation in this instance would have strengthened their case.

The loud objections by Senator Warren and others has served to bring daylight to the changes in financial regulation being proposed.  This will hopefully make it more difficult to push through further, possibly much more damaging, changes to Dodd-Frank at the behest of the banks.

Ensuring attention was paid to the issue also served to make clear who in the House and the Senate are in fact in favor of bank bailouts.  Weakening Dodd-Frank will increase the likelihood (even if only marginally so) that bank bailouts will be necessary in some future crisis to protect FDIC insured deposits and to protect the economy from a full financial collapse.  Republicans, including in particular Tea Party supported Republicans, have asserted they are against bank bailouts.  But their actions here, with the Dodd-Frank amendments inserted into the budget bill at the insistence of the House Republican leadership, belies that.

The actual economic substance of the Dodd-Frank amendments might therefore be limited, especially if there is now the regulatory response that should be required in the environment of certain banks holding more risky assets.  But the politics may be quite different, and could explain why there was such a vociferous response by Senator Warren and others to this ultimately successful effort to weaken a provision in Dodd-Frank.

The Impact of Austerity Policies on Unemployment: The Contrast Between the Eurozone and the US

Unemployment Rates - Eurozone and US, Jan 2006 to Oct 2014

A recent post on this blog looked at the disappointing growth in the Eurozone since early 2011, when Europe shifted to austerity policies from its previous focus on recovery from the 2008 economic and financial collapse.  There has indeed been no growth at all in the Eurozone in the three and a half years since that policy shift, with GDP at first falling by about 1 1/2% (leading to a double-dip recession) and then recovering by only that same amount thus far.  The recovery has been exceedingly slow, and prospects remain poor.

The consequences of the shift to austerity can be seen even more clearly in the unemployment figures.  See the chart above (the data comes from Eurostat).  Unemployment in Europe rose sharply starting in early 2008 and into early 2009.  But it then started to level off in late 2009 and early 2010 following the stimulus programs and aggressive central bank programs launched in late 2008.  Unemployment in the US followed a similar path during this period, and for similar reasons.

But the paths then diverged.  After peaking in early 2010 at about 10% and then starting to come down, the unemployment rate in the Eurozone switched directions and started to rise again in mid-2011.  It reached 12.0% in early 2013 and has since come down slowly and only modestly to a still high 11.5% currently.  In the US, in contrast, the unemployment rate reached a peak of 10.0% in October 2009, and has since fallen more or less steadily (with bumps along the way) to the current 5.8% (as of October 2014).  It has been a slow recovery, but at least it has been a recovery.

This divergence began in 2010, as Europe shifted from its previous expansionary stance to austerity.  Influential Europeans, in particular German officials and Jean-Claude Trichet (then the head of the European Central Bank) argued that not only was austerity needed, but that austerity would be expansionary rather than contractionary.  We now see that that was certainly not the case:  GDP fell and unemployment rose.

The most clear mark of that shift in policy can be found in the actions of the European Central Bank.  ECB interest rates had been kept at a low 0.25% for its Deposit Facility rate (one of its main policy rates) for two years until April 2011.  The ECB then raised the rate to 0.50% on April 13, and to 0.75% on July 13, 2011.  But European growth was already faltering (for a variety of reasons), and it was soon recognized by most that the hike in ECB interest rates had been a major mistake.  Trichet left office at the end on his term on November 1, replaced by Mario Draghi.  On November 9 the ECB Board approved a reversal.  The Deposit Facility rate was cut to 0.50% that day, to 0.25% a month later on December 11, and to 0.00% on July 11, 2012.

Fiscal policy had also been modestly expansionary up to 2010, as monetary policy had been up to that point, but then went into reverse.  Unfortunately, and unlike the quick recognition that raising central bank interest rates had been a mistake, fiscal expenditures have continued to be cut since mid-2010.

Germany in particular called for cuts in fiscal spending for the members of the Eurozone, and forced through a significantly stricter set of rules for fiscal deficits and public debt to GDP ratios for Eurozone members.  Discussions began in 2010, amendments to the existing “Stability and Growth Pact” were approved on March 11, 2011, and a formal new treaty among Eurozone members was signed on March 2, 2012.  The new treaty (commonly referred to as the Fiscal Compact) mandated a balanced budget in structural terms (defined as not exceeding 0.5% of GDP when the economy was close to full employment, with a separate requirement of the deficit never exceeding 3% of GDP no matter how depressed the economy might be).  Financial penalties would be imposed on countries not meeting the requirements.

The result was cuts to fiscal expenditures:

Govt Expenditures, Real Terms - Eurozone and US, 2006Q1 to 2014 Q2 or Q3

Government fiscal expenditures in the Eurozone had been growing in real terms in line with real GDP up to 2008, at around 2 to 3% a year.  With the onset of the crisis, fiscal expenditures at first grew to counter the fall GDP.  But instead of then allowing fiscal expenditures to continue to grow even at historical rates, much less the higher rates that would have been warranted to offset the fall in private demand during the crisis, fiscal expenditures peaked in mid-2010 and were then cut back.  By 2014 they were on the order of 14 to 15% below where they would have been had they been allowed to keep to their historical path.  This has suppressed demand and therefore output.

The path of US real government expenditures is also shown on the graph.  Note that government expenditures here include all levels of government (federal, state, and local), and include all government expenditures including transfers (such as for Social Security).  Government expenditures for the Eurozone are defined similarly.  The US data comes from the BEA, while the Eurozone data comes from Eurostat.

Government expenditures in the US also peaked in 2010, as they had in the Eurozone, and then fell.  This has been discussed in previous posts on this blog.  But while US government expenditures fell after 2010, they had grown by relatively more in the period leading up to 2010 than they had in the Eurozone, and then fell by relatively less.  They have now in 2014 started to pick up, mostly as a consequence of the budget deal reached last year between Congress and President Obama.  State and local government expenditures, which had been severely cut back before, have also now stabilized and started to grow as tax revenues have begun to recover from the downturn.  And in part as a result, recent GDP growth in the US has been good, with real GDP growing by 4.6% in the second quarter of 2014 and by 3.9% in the third quarter.

The fiscal path followed in the US could have been better.  An earlier post on this blog calculated that GDP would have returned to its full employment level by 2013 if government spending had been allowed to grow merely at its historical rate.  And the US could have returned to full employment by late 2011 or early 2012 if government spending had been allowed to grow at the more rapid rate that it had under Reagan.

But with the fiscal cuts, unemployment has come down only slowly in the US.  The recovery has been the slowest of any in the US for at least 40 years, and fiscal drag by itself can account for it.  But at least unemployment has come down in the US, in contrast to the path seen in Europe.