Fed Chairman Ben Bernanke's recent pronouncements were confusing to everyone. He suggested he might be ready to pull the plug on policies that are no longer necessary because they worked so well, when in fact, they haven't worked at all and cannot be expected to. Do you sell because he's half right, or because he's half wrong?
Let's just look at the labor market. We have achieved a normal cyclical recovery that feels absolutely awful because U.S. Labor Market suffers from deepening secular and structural issues that cannot be solved with either monetary policies or indirect fiscal stimulus. Still, those calling for Bernanke's head because they think these structural problems render his musing on an end of quantitative easing premature may have been missing the point of QE all along: QE can't solve structural problems. To the extent that anyone tries to use stimulus to resolve structural problems, all you get is unwanted inflation. Presuming Ben Bernanke knows this, he'd want to find a way to exit QE long before the stated goals of QE are ever reached because they can't be reached this way. He would also want the markets to be confused so that the unwanted side-effects of over-priced asset markets never get too far out of hand. Perhaps his plan is working exactly as he expected.
I don't have any idea what is inside the little man's head, but I can show that we have serious structural problems and why they can't be addressed with stimulus. We cannot leave everything to the Fed. In this analysis, I'm going to concentrate on the economy's ability to create jobs for those who are seeking jobs. I am aware that the participation rate, which is the percentage of people in the working-age cohort that are seeking work, has declined, but we need to first provide work for those who want it, and we've failed miserably even at this. The actual participation rate stands at 63.4%, which compares to a peak of 67.3% in April of 2000. That is a rate of decline of only 0.3% per year and only about half of the decline is attributable to the current recession.
In the last 12 months, the U.S. economy has created an average of 150,000 jobs per month. The number of people seeking work, however, has risen by 50,000 per month, so the job creation is only exceeding the growth in labor supply by 100,000 per month. If we plot this net job creation on cumulative basis, as shown in chart 1 below, we can see several aspects of the current labor market come in to much sharper focus:
1. Nearly 10 million needed jobs were lost between 2007 and 2009.
2. Before that, more than 7 million needed jobs were lost in the 2000 recession, and only about a million of these were ever recovered.
3. The pace of recovery in the past 4 years has been as fast as any previous recovery, but it is neither sufficient nor does it appear to be sustainable.
This last observation is based on a mere projection of current trends. The number of new jobs filled currently exceeds the number of new job seekers, but the number of new job seekers is growing at a faster rate. This isn't a forecast, it is just an extrapolation of where we are headed if nothing changes.
If we use a similar methodology to compare previous recessions, we can learn even more about why this recovery seems so much more painful than others before it. In the charts below, I've indexed job creation relative to the index in available work-force. What stands out to me in this chart is neither the depth of the current recession nor the slowness of its recovery, but rather, that we never recovered to full employment from the previous recession 10 years earlier. What makes this recession so horrible is that we started from the place where most other recessions were at their worst.
This is not all. One final aggregate piece of data says we are not on the path to healing. Today, four years after the recession began, the ratio of part-time to full-time jobs ratio is still lower than it was at the nadir of most previous recessions.
What's the Solution?
Monetary policy can have some effect on cyclical issues in the labor market, but most of the problems we face are clearly structural. This gets clearer when we look at the trend in specific industries. Two industries, construction and manufacturing accounted for 50% of the job losses in this recession.
Construction formed one of the unique features of this recession. Although always a deeply cyclical industry, it has always employed a very consistent average of about 4.2% of American workers over any complete cycle. This recession was very different because the housing bubble led the sector to a peak of 5.1% of the workforce, and later falling to a trough of 3.8%, where it has hardly changed for the past 3 years. The worst previous down turn was in 1973, where the sector fell from 4.7% to 3.8% of the workforce.
Although it would probably not be desirable to return to the previous peak levels, this is an industry where direct and concentrated stimulus could be extremely effective in getting this industry back to average levels that prevailed before the bubble. Doing so would reduce the unemployment rate by a full percentage point. We would have to be careful not to over-stimulate the housing industry which still suffers from some inventory over-hang, but there is no place in America that is more crying out for increased investment than our dilapidated infrastructure. Those who would make the argument that we don't have enough money should consider how much more it will cost if we wait until we are forced to fix these infrastructure related problems later when either interest rates or the cost of materials or both are much higher.
Manufacturing forms the second largest problem in this recession, but this problem is obviously more structural and can't be fixed with either monetary or fiscal stimulus. In every single recession, manufacturing jobs have been lost at a greater pace than non-manufacturing jobs, and almost none of these jobs were ever recovered. Manufacturing jobs accounted for 25.8% of the work-force in 1953, but have dropped to less than 8% today.
Some have even argued that since manufacturing is only 8% of the workforce, nothing we can do would have a large enough impact on employment. This analysis fails to understand the significance and the unprecedented recent collapse in manufacturing jobs. This industry alone accounted for 25% of the job losses in the current recession, even though it was not associated in any way with the preceding bubble. In fact, manufacturing jobs dropped to 10% of the workforce in the previous recession, from 12.8%. They then proceeded to drop another 1.1%, to 8.9% during the bubble!. Adding insult to injury, they finally dropped to 7.7% in the current recession.
There is widespread agreement that manufacturing is important source of high-quality jobs, but very little thought seems to have gone into what the optimal ratio of manufacturing to service jobs might be. A 2010 department of commerce report found that the average manufacturing job offered a 17% premium in hourly compensation over non-manufacturing jobs. This was generally true at every level of education. There is of course, considerable diversity. Not all of the good jobs are in manufacturing and not all of the bad jobs are in the service sector.
But it is not the good jobs that we are losing. America isn't losing its manufacturing strengths. Throughout this period, the percentage share of GDP contributed by manufacturing output has remained fairly constant at about 15%. The carnage in manufacturing jobs has occurred because our manufacturers keep finding ways to produce more with less labor. American manufacturers generate more profit today than at any time in history. We have improved the profitability, however, without considering what the heck it is that we are trying to make the profits for. What purpose do these profits serve if there are not enough jobs for our workforce?
If you regularly pay multi-million dollar bonuses to executives who are able to ruin thousands of lives with the flick of a pen, then people who can do this kind of work without remorse will come out of the woodworks. I am fairly certain that if managers had their pay docked for every layoff, rather than boosted, we would never have had this problem in the first place.
A more practical solution may be to insert circuit breakers, just as we have done in the stock market, to allow the free market more time to adjust to the shocks caused by lay-offs. In my view, companies need to be able to locate their production where it is most economic for them to do so, and we do not want the government to get involved in deciding which jobs should go where. But we need to give the economy more time to heal from these shocks. All it would take is a minimal set of inverse incentives to make companies at least think twice before they ship jobs overseas. Simple tax incentives that make layoffs less economically attractive for profitable companies would probably make a significant difference. This would still allow companies to shed staff when they can demonstrate that their survival is at stake, but wouldn't allow managers to reap multi-million-dollar bonuses for spurious staff reductions and it just might encourage companies to seek profitable ways to redeploy displaced workers within their own business, and prevent them from relying on layoffs to enhance profits that already provide an adequate return to shareholders. This kind of policy could stimulate demand and innovation, and would not involve the government in deciding who wins and who loses.
This is only one of many possible ideas. Regardless of the specific policies chosen, we have passed the point at which we can no longer afford to keep ignoring this problem. It has become more than obvious that the problem is not solving itself in any way that could be considered desirable. Another distinguishing feature of the current recession is that it is the first ever in which non-manufacturing job losses were significant, and in fact, today, four years after the start of the recession, the non-manufacturing jobs are still lower relative to job seekers than at the deepest point of most previous recessions. Non-manufacturing job growth is no longer going to be capable of covering up the manufacturing job problem that we have in this country.
The Retail industry is a perfect example of the brewing trouble in non-manufacturing. Retail accounted for third largest number of job losses in the current recession, and apparently this has become more of a structural decline than a cyclical one. In every past recession, retail dropped less than a percentage of its share of the total workforce and then quickly recovered to new highs. But during the housing bubble, retail employment never recovered to its pre-2000 high, and is not keeping pace with general industry employment trends in the current recovery either. Increased share of on-line retailing, as well as consumers growing acceptance of self-service check-outs are apparently driving this trend.
This appears to be a natural process of creative destruction that we need to just let happen, but we could slow it down during the next recession just by enacting the same tax incentives to discourage lay-offs. This would not prevent the inevitable, nor would we want it to, but it would give us time to find replacements for those jobs.
Financial services formed the fourth largest sector for job losses, and although much smaller in size compared to retail, these jobs are generally much higher paying. Like retail, this sector's share had risen in every previous expansion, but peaked in 2000. It suffered little decline at first, as it surely benefited from the housing bubble, but has failed to show any life in the current recovery. It's too early to be certain, but if we assume the effects of the housing bubble masked an already underway structural decline in this segment, then here too, is an industry where we cannot expect very much if any gains - no matter what we do policy wise.
Wholesale trade suffered the fifth largest job losses, and is also a non-manufacturing industry in crystal clear secular decline. Since this industry comprises a very diverse group of businesses, it's hard to get a handle on what the cause is, but because it is a distribution business, it seems doubtful that it is caused by shipping jobs overseas. More likely, it is a combination of technological advances and increased concentration of the industry into a smaller number of employers. At 3.6% of the total, though, it sees less unrealistic to hope that most of the damage has already been done and that there isn't much to be gained by focusing on this industry.
State and Local Government hiring caused the sixth largest decline in jobs. The actual percentage is small: governments have cut their share of the work force to 12.3% from 12.8%, but at a time when hundreds of thousands of people are out of work, the federal government needs to step in and provide assistance so that states and local governments can expand, not contract. Raising the share of the workforce to 13.6% would reduce unemployment by more than 1%, while adding only 0.7% to the federal debt assuming total compensation per new government employee were 50,000. There is not a single government agency anywhere in this country that is getting things done as fast as they could be, and to suggest that this many people could not be gainfully employed or that this would not be a worthwhile expenditure is pure insanity.
Not in America
In the United States, we don't like to believe that governments have any more than an indirect role in creating jobs, but for 10 million people, the invisible hand has failed, and there seems to me to be no excuse for this astonishing tragedy. We could have prevented this recession from ever becoming so deep. We can still very quickly rejuvenate the labor market if we choose to.
Sadly, rather than look at the problem and try to address the specific issues, the approach of our politicians, which reflects the approach of our citizens, is to look at unemployment situation as a morality play. We use the numbers not to analyze, but to win arguments.
We could easily reduce the unemployment to 6% with the government and construction job stimulus programs I suggested. A more pro-active approach to manufacturing jobs could easily cut the rate to 4% over time.
I have tried to discuss this issue without the slightest hint of whose political party I side with, except that I believe it is self-evident that the government has an inherent responsibility to achieve full employment on a consistent basis. Sadly, I know that the comments that follow from SA readers will not be so disciplined. If anyone feels compelled to bring any of the tired and frankly uninteresting political dogma into this debate, please have a look at my final chart, which is the same as the first chart except that I have colored in blue the parts over which democratic administrations presided, and in red the parts which republicans presided. Then consider the following:
1. Is the presiding president really responsible for the declines that occurred at the beginning of his tenure but that started well before him? If so, Ronald Reagan and Barack Obama both lost almost exactly as many jobs in the first part of their tenure, and so far, Obama's recovery looks exactly the same as Reagan's.
2. George Bush, a republican, presided over the worst job performance in recorded history, but was that really all his fault, or did his predecessor set him up by creating and unprecedented number of unsustainable jobs?
3. Bill Clinton had by far the best record of job creation of any president that we have data for, but did he really do that, or did he just ride the headwind that was created by the trend toward deregulation initiated by Reagan?
4. Nixon, and Ford, both republicans, presided over the second worst job performance, but their situation was not unlike Bush's. Could they have been set up by the unsound expansion of the Kennedy era?
5. Is there any sign anywhere in the data, that any particular healthcare policy had any noticeable impact on the trend in employment? Actually, the job recovery began just about the time of the affordable healthcare act was passed. Recent research actually suggests that Obamacare is more likely to reduce total compensation costs, but for current hiring, it matters what employers think more than what is real. I have seen plenty of diatribes on what Republicans think the cost will be, but nothing on what actual employers think the cost will be. We can make some common sense assumptions though. It is logical that anything increasing the cost of labor would cause a decline in hiring, but healthcare costs are about 17% of total labor costs in the U.S. today. They have been rising at a rate of 6% annually for some time, so if costs rise 10% after Obamacare, this only amounts to a 4% change in a 17% item, i.e., about a total impact on labor costs of much less than even 1%. Getting rid of Obamacare would not solve our problem.
6. Is there any sign in the data that any specific pro-regulatory or anti-regulatory political regime shift caused any immediate change in employment trends. It makes sense that there is a cost/benefit trade-off for any regulatory change, but I don't see any confirmation of this in the data. Perhaps it is because there really haven't' been that many important regulatory changes that are relevant to employment as the proponents of this theory claim.