The Biggest Cut in Unemployment Benefits

Copyright, The New York Times Company

While Congress has been debating whether to cut the duration of unemployment benefits, perhaps the largest unemployment benefit cut occurred when the stimulus law expired.

Unemployment insurance offers funds, for a limited eligibility period, to people who lost their jobs and have not fyet been able to find and start a new job. In 2008, “emergency unemployment” legislation, plus automatic triggers in the unemployment insurance rules, extended the eligibility period to up to 99 weeks from 26 weeks.

Several times since then, and as recently as last week, new legislation has prevented the eligibility period from returning to 26 weeks.

The length of the eligibility period has received much attention; it affects how much the program spends and how much unemployed people receive. For example, if the weekly benefit were $275, and an unemployed person were unemployed for a year, then the average weekly benefit he would receive under the 26-week rule would be about $138 ($275 for half the year, and zero for the other half).

By extending the eligibility period to more than 52 weeks, this person would see his average weekly benefit increase to $275 from $138.

The green line in the chart below shows the average weekly benefit received by unemployed people over time, assuming that:

(a) they were receiving $275 a week until their benefits were exhausted
(b) about half of the aggregate time unemployed occurs in the first 26 weeks
(c) essentially all unemployment spells end in less than 99 weeks

These assumptions are a close approximation to the unemployment spells experienced by people 25 to 64 during 2010. For the reasons explained above, the line jumps to $275 from $138 in mid-2008, and then is constant thereafter.

However, the eligibility period is not the only part of the unemployment insurance rules that have changed since the recession began. The American Reinvestment and Recovery Act (the “stimulus law”) made a number of additional changes.

It increased the weekly benefit by $25 a week (and guaranteed that the $25 increase would not cause anyone to lose Medicaid coverage); federally funded 100 percent of extended benefits; exempted the first $2,400 of unemployment insurance received in 2009 from federal income tax; and paid 65 percent of an unemployed person’s health insurance premiums.

The act also paid states about $7 billion to allow more of the unemployed to qualify for benefits.

The federal financng of extended benefits meant that employers would not be liable for the extended benefits received by their former employees, which makes it less profitable for them to contest unemployment claims made by their former employees.

A $25 weekly benefit bonus was clearly worth $25 a week for as long as it lasted (until mid-2010). At a marginal federal income tax rate of 21 percent, the exemption from federal income tax on the first $2,400 of unemployment insurance received in 2009 is worth about another $10 a week.

Perhaps the most valuable added benefit was the health insurance subsidy. For unemployed people who, through the Cobra program, continued to participate in the health insurance plan they had with their former employer, the federal government would pay 65 percent of the premium. For such people, this subsidy is estimated to be worth about $170 weekly. This benefit ended in mid-2010.

The red line in the chart shows the combined unemployment benefits for an unemployed person participating in the Cobra program, excluding any benefits received from other safety-net programs such as food stamps or Medicaid.

The weekly benefit peaks in 2009 at $455. The increase in early 2009 when the stimulus law passed is even greater than the increase in mid-2008 from the lengthening of the eligibility period.

It is not yet known how many unemployed people received the Cobra health insurance subsidy, but many who did not had their health insurance covered by another federal program, Medicaid.

Moreover, a $455 weekly benefit is not small change; it is much more than someone would earn on a full-time job that paid minimum wage. Among the 106 million working-age heads of households and their spouses lucky enough to be working in 2009, about 25 million of them were earning less than $455 a week.

Even though Congress has not yet let emergency unemployment benefits expire, the largest unemployment benefit cut may have already occurred in 2010 when the stimulus law expired.

Productivity and Real wages since 2007

A number of bloggers have recently discovered real wages as a labor market indicator. They are at least 3 years late to the party.

Three years ago I blogged about the effect of labor supply on real wages.

I noted how real wages had risen since 2007, and predicted that they would begin to decline in 2010.

I have continued to update this work, eg here, and here.

The fact is that the real wage time series fits my recession narrative very well.

Unemployment Compensation Over Time

Copyright, The New York Times Company

The propensity of unemployed people to receive unemployment benefits reached historical highs after the 2008-9 recession and may indicate that benefit rules have more impact on the economy than ever before. The changing aggregate impact of unemployment insurance may be worth considering as Congress debates benefit extensions.

Unemployment insurance offers funds, for a limited eligibility period (now up to 99 weeks), to “covered” people who lost their jobs and have yet been unable to find and start a new job.

Some economists suggest that unemployment insurance prolongs unemployment because recipients have to give up their benefits as soon as they find and start a new job, or return to working at a previous job.

Some economists also say they believe that unemployment insurance stimulates spending because unemployed people are thought to spend most, if not all, of the money they have on hand.

But neither of these effects can operate unless people take part in the program.

Historically, many of the jobless have not collected unemployment benefits because of ineligibility, lack of awareness or unwillingness to do so.

The chart below graphs the recipiency rate — the percentage of people unemployed who are collecting unemployment benefits that week — to 1986. It was calculated from weekly data, then averaged over 52 weeks to remove some of the large seasonal patterns.

The percentage is always well under 100, fluctuating from 31 to 68 percent. The peak recipiency rates seem to follow recessions; three national recessions have occurred since 1986, in 1990-91, 2001 and 2008-9. Previous studies covering the period 1960-94 found a similar pattern (although perhaps no recipiency rate peak was found after the 1981-82 recession), with a maximum recipiency rate for all 35 years of about 50 percent.

Some unemployed people cannot collect benefits because they quit their jobs, rather than being laid off. But quits are less common during recessions, one reason the recipiency rate is greatest during recessions.

Another reason that recessions can have high recipiency rates is that, by law, benefit eligibility periods are longer during recessions. Laws often increase the eligibility periods by a greater percentage than the average duration of unemployment increases, with the result that a larger percentage of the unemployed are eligible for benefits.

Among other things, the 2009 American Reinvestment and Reinvestment Act expanded eligibility for unemployment insurance by encouraging states to adopt an “alternative base period” benefit calculation rule that allowed a number of people with weak employment histories to qualify for benefits.

Long-term comparisons of recipiency rates are tricky because the data sources change and because of secular changes in the composition of unemployed, but it appears that recipiency rates were higher during 2009 than they have been in 50 years, and perhaps ever.

With such a large percentage of unemployed people receiving benefits, the potential employment and spending effects of those benefits may be greater than ever.

Hiring, ‘Quits’ and Layoffs

Copyright, The New York Times Company

The recession and lack of recovery have often been characterized as a lack of hiring, rather than an extraordinary number of employer-employee “separations” (a separation is a layoff or what the Bureau of Labor Statistics calls a “quit”). But the aggregate data on hiring, quits and layoffs can be misleading.

The chart below from the Bureau of Labor Statistics shows total hires, separations and total employment from December 2000 to December 2009. The shaded area to the right indicates the start of the most recent recession. Both separations and hires fell about 20 percent (the hires dropped sooner).

Bureau of Labor Statistics

However, this characterization is quite different for young people than for the rest of the work force. Even during business cycle expansions, young people have high rates of job turnover. The number of young people hired during a typical month is disproportionately high, as is the number of young people quitting or getting laid off (I use the term “layoff” to refer to both layoffs and discharges).

For this reason, young people could dominate the job turnover statistics, even while they do not dominate the employment statistics.

A paper by Michael Elsby, Bart Hobijn and Aysegul Sahin, using a method developed by my University of Chicago colleague Robert Shimer, estimated job separations for different age groups according to the number of people flowing into unemployment. They found a very different pattern for people 16-24 than they did for people 25-54.

Estimated job separations among employees ages 25-54 were 33 percent greater in 2009 than they were in 2007. I tried to update their estimates through 2010 and found that job separations still remained greater than they were in 2007.

In contrast, the low employment rates for young people since 2007 are almost entirely explained by low hiring rates.

The Bureau of Labor Statistics also reports that the composition of job separations has changed a lot since 2007. Before the recession began, quits were by far the most common type of separation; now the number of quits about equals the number of layoffs.

Perhaps the decline in quits is a signal of what’s ailing the economy, although I view it largely as a consequence of the unemployment insurance system. A person who quits his or her job is not eligible for unemployment insurance. As a result, calling a job separation a “quit” rather than a “layoff” results in the loss of unemployment benefits.

I estimate that, before the recession began, a person beginning unemployment would receive about 12 weeks of unemployment benefits, on average, if he or she were eligible for benefits. By 2010, that average was up to about 34 weeks. At $300 a week, that means that the government subsidy for calling a separation a “layoff” rather than a resignation was up to about $10,000 from $3,600.

There is also the employer payroll tax consequence of layoffs to consider, but overall employers and employees now often have a lot to gain by calling their separation a layoff rather than a quit.

These are a couple of reasons the aggregate data on hiring, quits and layoffs can be misleading. Even so, the very different patterns for people 16-24 and 25-54 may suggest that the recession and lack of recovery have more than one cause.

Species Protection and Technology

Copyright, The New York Times Company

Advances in genetics may change aspects of environmental policy in the future.

Protecting endangered species is a legitimate goal for public policy. But we cannot expect that public policy will prevent all species extinctions, because scientists estimate that hundreds of species, if not thousands, become extinct every year, many without our knowledge.

A rational approach to protecting endangered species would balance costs and benefits. Landowners bear many of the costs, because they can lose title to land that is home to an endangered species or be restricted in how they can use their land. Businesses bear some costs, too, as they are sometimes required to alter their activities to protect an endangered species.

Our planet benefits from biodiversity, and there can be a large “option value” to helping an endangered species live another generation. We can change our minds at some later date and no longer try to prevent a species’ extinction. But it would seem impossible to change our minds in the other direction — once a species is gone, it would seem too late to decide that we wish we had protected it.

But technology may be changing the option-value calculation, because scientists are learning how to clone extinct animals. The time will come when scientists will produce living versions of previously extinct animals. With enough time, they would probably be able to do so cheaply.

To the extent that cloning will someday be possible, the option value of preserving an endangered species is a lot less. In some cases, it may be cheaper to save some DNA, and let a future, richer and perhaps more enthusiastic generation make its own copy of the species.

One objection to the cloning approach is that cloning is itself too expensive. That’s true with today’s technology, but future genetic technologies will invariably be more efficient (cheaper for what they produce).

Another objection is that a cloned species would not be genetically diverse enough to survive, because it would be cloned from just a few DNA samples. But that objection also springs from today’s technology. Scientists may learn to save a genetically diverse DNA sample or even to produce genetic diversity themselves.

Yet another objection is whether we would have room in the future to house a cloned species. But it would seem to me more space intensive to protect each endangered species’ habitat for hundreds of consecutive years than it would to store some DNA and find space when the time comes for cloning.

The fact is that species go extinct all of the time, while we incur real economic costs protecting others. By bringing cloning into the set of public policy instruments, we can protect more species, reduce economic costs of protection, or both.

Bankers, Too, Cast a Safety Net

Copyright, The New York Times Company

The combination of housing market events and the profit motive of mortgage lenders turned trillions of dollars of household debt into a huge safety net.

Household debt had been increasing during the 1980s and 1990s, but the rate of increase was extraordinary in the years leading up to the recession. By 2007, household sector debt had reached 114 percent of the nation’s personal income – more than $14 trillion. The change was almost entirely due to accumulation of home mortgage debt.

Normally, home mortgages are fully secured by a residential property, and when a homeowner fails to make the scheduled payments on time, the lender can seize the property and sell it to recover its principal, interest and fees. When the lender has this valuable foreclosure option, borrowers overwhelming either make their home mortgage payments on time or sell their property in an orderly fashion to obtain the money to repay the mortgage lender, even in cases when the homeowner is unemployed.

When residential property values plummeted in 2008 and 2009, a number of residential properties were suddenly “under water” — worth less than the mortgages they secured. In those cases, the lender’s foreclosure option was no longer valuable – selling the property would be likely to yield too little money to cover principal, let alone interest and fees.

Lenders needed a way to estimate which borrowers would still pay in full and a way for other borrowers to work out a mortgage modification that would give them an incentive to pay at least a bit more than their homes were worth.

Naturally, a borrower’s income is a factor considered – borrowers with high income can be expected to repay more than borrowers with low income. Thus, a partial solution to the lenders’ collection problem is to insist that high-income borrowers pay more of the mortgage amount due and allow at least some low-income borrowers to pay less.

From this perspective, the lenders’ desire to maximize debt collections (after the collapse of residential real estate values) causes them to create a kind of safety net program that gives low-income people more help with their housing expenses (much the way the federal food stamp program gives low-income people more help with their food expenses) in the form of modified mortgage payments.

To quantify the size of the loan modification safety net and its changes over time, I estimate the amounts that “home retention actions” (as the federal government calls these mortgage modifications that allows people to stay in their homes) actually changed mortgage payments from the original mortgage contract, which specified only payment in full or foreclosure.

To estimate those amounts for 2008-10, I first measured the number of residential properties in each quarter receiving loan modifications, lender permission for short sale or lender permission for deed-in-lieu of foreclosure.

Next, I multiplied the number of transactions by a $20,319 average value of each loan modification (a typical modification reduced monthly payments by $400 for a minimum of 60 months; at an annual discount rate of 7 percent, that’s a present value of $20,319). I do not have data on the number of home retention actions for the years 2006 and 2007, but I assume the dollar value of discharges those years were, as a proportion to discharges in 2008, the same as total mortgage loan discharges by commercial banks.

Because the home retention actions are necessary primarily when homes are worth less than the mortgages they secure, the amount discharged by home retention actions is much less in 2006 and 2007 when residential property values were still high. During 2010, mortgage lenders discharged more than $70 billion of mortgage debt through home retention actions. Seventy billion dollars for one year is small in comparison to the total amount that homeowners were under water but is more than the spending by the entire food stamp program for that year.

The last row of the table displays discharges on other consumer loans, such as credit card debt. Those discharges are smoother over time because they are not directly tied to the housing cycle but still totaled more than $70 billion in 2010. The combination of discharges of other consumer loans and discharges of home mortgages by home retention actions was almost $150 billion in 2010, which exceeds the peak spending for entire unemployment insurance system.

Bankers deserve a lot of blame for getting us into this mess, have dipped far too deeply into the United States Treasury to help themselves, and have been far too slow to modify mortgages. For these reasons, it’s remarkable that their own selfish pursuits have forced them to create a safety net of sorts that rivals the amounts spent by public sector safety net programs.

Income Security Became More Generous

Professor Krugman is reiterating his claim that "income security" programs grew solely because of the recession. He is incorrect: only a small fraction of that growth is due to the recession, the rest because of legislation making the programs more generous.

Inflation-adjusted government spending on unemployment insurance (UI) and food stamps (SNAP) has more than doubled on a per capita basis, and most of this growth is due to changes in eligibility rules, and increases in payments per eligible person, rather than increases in the number of people who would have been eligible under pre-recession program rules.

By 2009, the UI program was absorbing a larger fraction of earnings lost due to unemployment than it did in 2007, with the majority of its expenditure made pursuant to federal expansions since 2007. Indeed, among persons aged 25 and over, it was more common to experience unemployment without government help before the recession than it was since 2007.

See also
How Unemployment Benefits Became Twice as Generous
UI Takeup Increases Sharply during the Recession
The Sharp Increase in Food Stamp Spending
Millions Caught by the Social Safety Net

Perhaps legislation changes were the "right thing to do" given the recession. But Professor Krugman is pretending that the legislation changes did not occur.

Millions Caught by the Social Safety Net

Copyright, The New York Times Company

Despite the severe recession, relatively few people saw their living standards fall into poverty, thanks to the social safety net.

According to the Center for Budget and Policy Priorities, the percentage of the United States population living in poverty increased by 0.6, to 15.5 in 2010 from 14.9 in 2007.

The poverty measure refers to resources available to families, accounting for the taxes they pay and subsidies they receive. Considering all that happened in the economy over those three years, 0.6 percentage points is quite a small change. Measures of the poverty rate typically change more than that over any three-year interval.

The study found that many people were technically above the poverty line in 2010, although their incomes were low, because they received government assistance like unemployment insurance, food stamps and refundable tax credits. The government assistance permitted them to have living standards above poverty, even while their market incomes were below the poverty line.

Were it not for government assistance, the study found, the recession would have pushed 4.2 percent of the population into poverty, rather than 0.6 percent.

One interpretation of these results is that the safety net did a great job: For every seven people who would have fallen into poverty, the social safety net caught six. Perhaps if the 2009 stimulus law had been a little bigger or a little more oriented to safety-net programs, all seven would have been caught.

Another interpretation is that the safety net has taken away incentives and serves as a penalty for earning incomes above the poverty line. For every seven persons who let their market income fall below the poverty line, only one of them will have to bear the consequence of a poverty living standard. The other six will have a living standard above poverty.

The safety net was not as effective before the recession began. As I explained in my last two posts, government assistance programs have not only supported more people but become more generous, thanks to changes in benefit rules since 2007.

Of course, most people work hard despite a generous safety net, and 140 million people are still working today. But in a labor force as big as ours, it takes only a small fraction of people who react to a generous safety net by working less to create millions of unemployed. I suspect that employment cannot return to pre-recession levels until safety-net generosity does, too.

The Sharp Increase in Food Stamp Spending

Copyright, The New York Times Company

The Department of Agriculture’s food stamp program, now known as the Supplemental Nutrition Assistance Program, or SNAP, provides money to low-income households for the purpose of buying food, often in conjunction with cash assistance programs. Adjusting for inflation, the program spent more than twice as much in 2010 as it did in 2007, before the recession began.

The Department of Agriculture found that the food-stamp spending increase “is likely attributable to the deterioration of the economy, expansions in SNAP eligibility, and continued outreach efforts.” Of particular relevance for the SNAP program is the fact that the poverty rate increased 18 percent, to 153 per thousand in 2010 from 130 per thousand Americans in 2007.

At least two eligibility expansions have occurred since the recession began: work requirements were lifted from April 1, 2009, through Sept. 30, 2010, and monthly income limits were 10 percent higher in the 2010 fiscal year than they were in the 2007 fiscal year, an increase about twice the rate of inflation over that period.

In addition, the American Recovery and Reinvestment Act increased maximum benefits by 13.6 percent, and the minimum benefit increased in October 2008. Increasingly, potential program participants have been given the opportunity to apply for benefits on the Internet.

The declining economy alone, under the previous rules, would have raised the spending on food stamps by 18 percent. But the revised provisions, enacted largely in response to the recession, are responsible for a greater share of the increase. The following table breaks down the program’s spending growth into three components: deterioration of the economy, relaxed eligibility rules and increased maximum benefits.

U.S. Department of Agriculture

The top row of the table is actual program spending for 2007 and 2010, adjusting for inflation and population. The second row of the table estimates the program’s hypothetical spending growth with 2007 eligibility rules, by assuming that real spending per capita increased since 2007 only in proportion to increases in the poverty rate, plus the 13.6 percent benefit increase of the American Recovery and Reinvestment Act. The last row assumes that real spending per capita increased only with the poverty rate. Under either scenario, the hypothetical spending increases are significant but well less than half of the actual spending increases.

Over all, the table suggests that most growth in spending on SNAP is due to changes in eligibility rules and increases in payments per eligible person. The program’s spending would certainly have grown if benefit rules had remained as they were in 2007, but much less than it actually did. And those more generous provisions are now likely to be here to stay, even if the conditions that prompted them abate.

UI Takeup Increases Sharply during the Recession

Copyright, The New York Times Company

It’s commonly assumed that unemployed people not receiving unemployment benefits have been unlucky enough to go without a job for so long that their benefits have run out. But often more important are limited work histories and a low propensity to take benefits that are available.

Historically, many unemployed people have not collected unemployment payments because of ineligibility, lack of awareness or simple unwillingness to collect benefits. But some of those patterns changed during the recent recession.

The chart below shows the number of unemployment compensation beneficiaries per unemployed person, for people 16 to 24, people 25 and over and all people 16 and over. This ratio can be less than one for all of the reasons mentioned and because some unemployed people may exhaust their benefits sometime during the calendar year.

Not surprisingly, more than three-quarters of young unemployed people do not receive unemployment compensation, in large part because they are much less likely to have the employment history that is required for eligibility. Young people are disproportionately represented among the unemployed, and their limited work histories are the primary reason why a large fraction of the unemployed does not receive benefits.

More striking is the increase to 85 percent from 50 percent among people 25 and over. Before the recession began, about a quarter of unemployed people that age had been unemployed for more than 26 weeks, when unemployment benefits were typically exhausted.

The remaining quarter of the unemployed did not receive benefits for a variety of other reasons: they may not have been interested in or aware of benefits, or they may have been ineligible because they quit their jobs (rather than lost them).

By 2010, unemployment was lasting much longer, but the time for receiving benefits had increased even more. Ninety-two weeks was a typical unemployment benefit period in 2010 (in some states it was 78 weeks, in others 99 weeks), yet only 12 percent of the unemployed 25 and over were unemployed that long.

That means as many as 88 percent of the people that age who were unemployed could have received benefits. That 85 percent received benefits tells us how rare it was for eligible people to forgo benefits during the recession.

The recipiency rate change from 2007 to 2010 is thus a combination of a decreased likelihood of exhausting benefits and an increased propensity to receive benefits early in the unemployment spell. These two factors change so much that even though the average weekly number of unemployed people 25 and over increased by more than six million from 2007 to 2009, the average weekly number of those people not receiving unemployment insurance actually fell by 700,000. (For the purposes of this calculation, I assume that, consistent with the law, nobody received unemployment benefits for a week that she or he was employed.)

This absolute decline in nonparticipating unemployed suggests that people are more willing (equivalently, less unwilling) to collect unemployment benefits than they were before the recession began.

Unemployment insurance is known for its ability to expand eligibility as a recession gets going, whether through the “extended benefits” that take effect at given jobless rates or through legislative action beyond that. But an adjustment almost as important has occurred in the labor force itself: during the recession, people increased their propensity to take advantage of available benefits.

How Unemployment Benefits Became Twice as Generous

Copyright, The New York Times Company

Government spending on unemployment insurance has soared, and it’s hard to imagine the program ever shrinking back to its prerecession size.

Unemployment insurance is jointly administered and financed by the federal and state governments, offering money to people who have lost their jobs and have as yet been unable to find and start a new job. On average they receive about $300 a week until they start working again, they stop looking for work or their benefits are exhausted.

Between 2006 and 2010, inflation-adjusted spending on unemployment compensation by federal, state and local governments more than tripled.

The program has been around for decades, but the most recent recession and continued economic weakness has created an especially large group of laid-off workers who, despite an extensive search, cannot find another job. More unemployed people equals more spending for the unemployment insurance program, so we expect the program to be spending a lot during a recession.

However, the unemployment program has also become more generous since 2006. Before the recession, an unemployed person in a state without high unemployment would often exhaust benefits after 26 weeks; that is, the program would stop paying after the 26th weekly benefit, even if the beneficiary was still without work.

The federal law in place before the recession included some local labor market “extended benefit” triggers that, based on the statewide unemployment rate, would automatically lengthen the maximum benefit period. These automatic triggers began to extend benefits around the nation in the middle of 2008.

About the same time, new “emergency unemployment compensation” legislation extended maximum benefit periods for the entire nation. The American Recovery and Reinvestment Act of February 2009 further extended these “emergency” periods to up to 99 weeks, and legislation later in 2009 and in 2010 permitted the 99-week maximum to continue. (Among other unemployment insurance expansions, the act also increased monthly benefit amounts and excluded from federal personal income taxation the first $2,400 of benefits received in 2009.)

The chart below shows the size of the “emergency” and extended-benefit expansions, by quarter, measured as a fraction of the entire unemployment insurance program. Essentially, no “emergency” and extended-benefit benefits were paid in 2007 or in the first half of 2008. “Emergency” and extended-benefit benefits immediately became about a quarter of all unemployment insurance benefits and beneficiaries and were a majority of all unemployment insurance benefits by the end of 2009 (the two measures are slightly different because they come from different data sources).

Because “emergency” and extended-benefit benefits are paid to people only when they have exhausted the normal benefits, the fraction shown in the chart is a measure of how much unemployment benefits are paid pursuant to unemployment insurance rule changes, as opposed to payments that occur merely because more people were losing their jobs.

If we assume, merely for simplicity, that the expansions had no effect on the number of people unemployed or on the length of time they were employed, then setting “emergency” and extended-benefit payments to zero as they were in 2007 would have cut total unemployment insurance benefit payments by the fraction shown in the chart. In this case, it appears that the unemployment insurance program is at least twice as generous as it was in 2007, thanks to the federal changes in benefit rules.

The unemployment insurance program is a good example of how federal government spending has grown and how tough it will be to bring it back to pre-recession levels. People are now used to having well more than one year’s unemployment benefits available to them, and politicians will have a lot of trouble asking them to make do with just 26 weeks.

Was Qaddafi Overpaid?

Copyright, The New York Times Company

By most measures, the former dictator Muammar el-Qaddafi looks to have been overpaid, even as dictators go.

Colonel Qaddafi’s wealth had recently been estimated in the tens of billions. But it now looks as though he could have been worth more than $200 billion.

Obviously, $200 billion, or even $10 billion, is a lot for wealth for one person. But $200 billion is but a fraction of Libya’s national wealth. Its proven oil reserves alone total 46 billion barrels. If those barrels were valued at $100 each, the oil reserves alone would be $4.6 trillion, or 23 times Colonel Qaddafi’s wealth.

In my research on dictators and their public finances, I estimate that, on average, dictators were taking about 3 percent of their nations’ incomes in the form of excessive taxation. Judging from Colonel Qaddafi’s share of Libya’s national wealth, that’s about what he was taking.

Dictators typically spend a lot on the military in order to protect themselves from people who might want to take their lucrative jobs, which itself is a sure sign that a dictator is overpaid. Led by Colonel Qaddafi, Libya’s government spent more of the nation’s income on the military than the average dictatorship does. Libya also spent less of its national income on social security than the typical dictatorship does, although perhaps a bit more than an economically and demographically similar democratic country would.

Colonel Qaddafi’s regime was known to torture and execute its political enemies. So it’s clear that the citizens of Libya were sacrificing too much for their leaders.

What’s less clear is whether the next leaders of Libya will take less or offer better services for the citizens of Libya. Egypt’s experience since Hosni Mubarak shows that the overthrow of a longtime dictator does not by itself bring freedom or democracy. Libya has much more in oil riches than Egypt, and political opponents in Libya are likely to find that wealth worth a violent fight.

Let’s hope that a long, bloody Libyan civil war does not make Colonel Qaddafi’s “fees” for his longtime leadership look cheap.

Three More Aggregate Hours Measures

On Wednesday I compared the household survey to the time dairy. Below are three more series, from the establishment survey.


The HOABS series is the one used by the BLS to measure productivity. In the 1982 recession, HOABS fell a lot less than AWHI:

Are Employers Requiring People to Work Longer Hours?

Copyright, The New York Times Company

As employers have sharply cut back employment since 2007, at least one survey asserted that existing employees have to work a lot more in order to maintain what was produced by the formerly larger work force.

The Census Bureau’s monthly household surveys do not suggest that such a pattern is widespread, because they measure that average weekly hours worked per employed person have fallen to 37.8 in 2009 from 39.0 in 2007. Another survey also measures hours worked, with a similar result. So it seems that the number of people employed and the hours they work have fallen, creating a huge drop in the economy’s total work hours.

But sometimes surveys can be misleading about hours worked, because people tend to report round numbers like “40 hours” or “35 hours” even when actual hours worked are not a round number (more than 40 percent of employed people in the monthly household survey reported that they worked 40 hours in the reference week, compared with a mere 0.4 percent who reported 39 hours of work). It is logically possible that a number of employed people were working more hours in recent years, but continued to report the round number of 40.

Since 2003, the Census Bureau has supplemented its population survey with the American Time Use Survey, dedicated to measuring time use. Participants in that survey are asked to account for all their waking hours in a specific day, listing various activities, including eating, watching television, working, traveling, caring for children and so on.

The diary study therefore has no bias toward finding that masses of people work exactly eight hours every day for exactly five days a week. It would be interesting to know if the recent recession looks different when the economy’s work hours are measured from the diaries, rather than from the population surveys as the product of employees and hours per employee.

The chart below displays the results. Eight calendar years are sampled, from 2003 to 2010. The blue line is based on the household survey and is an index (normalized to 100 in the year 2007) of the average number of hours worked by adults. It shows about a 2 percent increase in hours worked from 2003 to 2006. Hours worked were about the same in 2007 as in 2006. For each of the three years after 2007, work hours were significantly below the previous year.

The red line is also an index of hours worked per person — but based on the time diary methodology (here I look at the sum of hours spent at work and in “income-generating activities”). The time diary actually suggests there was a mild recession in 2004, because hours worked per person were lower that year than in the surrounding years. Also unlike the household survey, the time diary suggests that work hours in 2007 were abnormally high by comparison with all previous years.

The time diary closely agrees with the household survey measures for the years 2008-10, confirming that hours worked dropped sharply after 2007. Although a few employers may require their workers to work longer hours, the typical pattern since 2007 is fewer hours per employee, and fewer employees.

Growing Businesses Cut Payrolls, Too

Copyright, The New York Times Company

In many industries, sharp employment cuts during the recession cannot be attributed to a lack of demand.

The standard narrative of the 2008-9 recession and lack of recovery has been that the financial crisis, housing crash, excessive debt and other factors caused consumers to spend less, and businesses to invest less. With the private sector spending less, employers had a hard time selling their products, so they had to lay workers off, cut back on new hiring, or both.

As Paul Krugman put it, “Businesses aren’t hiring because of poor sales, period, end of story.”

Yes, consumer spending dropped sharply, as did business investment, in 2008 and 2009. But that observation does not tell us whether low employment is a result of low spending or if the reverse is true.

I agree that a few important industries, including manufacturing, home construction and much of the retail sector, did, and still do, suffer from significantly low demand. Those industries vividly illustrate the demand narrative — but they are only a minority of the overall private sector.

The lack-of-demand hypothesis is incorrect for a large fraction of the economy. The chart below illustrates output, revenue and employment from the United States wireless telecommunications industry (that is, cellphones). This industry has clearly not been suffering from a drop in customer demand.

Since 2007, the number of mobile connections has increased almost 20 percent, to 303 million from 255 million. The Bureau of Economic Analysis estimates that consumer spending on mobile communications increased 15 percent (not inflation adjusted) over that time frame.

Bureau of Economic Analysis

Despite continued demand growth, employers in the wireless telecommunications industry sharply cut employment, at an even greater rate than employers in other industries. After growing 6 percent from 2005 to 2007, the industry’s employment had fallen 14 percent by 2010.

There is no way to blame that sharp employment drop on “poor sales.”

This pattern is not limited to the cellphone industry. Other industries sharply cut back their employment even while their revenues were falling little, if at all; the employment loss from such industries numbers in the millions.

To examine this issue more systematically, I used the industry economic accounts published by the Bureau of Economic Analysis. Industries can be examined at varying levels of detail: I divided the private sector into 21 industries and classified them according to the percentage change in their revenue between 2007 and 2009. The table below shows the results.

Bureau of Economic Analysis

In two of the industries, education and health care, revenues grew more than 2 percent (in fact, their increase was about 10 percent), and their employment increased, as is shown in the table’s top row. Five other industries summarized in the next row had a revenue increase but still sharply cut their employment. Four others had minor revenue declines and cut their hiring sharply, too.

The number of full-time equivalent employees declined 2.2 million in those nine industries combined, even though it seems that those industries had enough sales to maintain their employment. Something else motivated them to cut employment and motivated them to forgo an opportunity to hire some of the many workers laid off by declining industries.

As I wrote before much of the employment decline happened, I think “some employees face financial incentives that encourage them not to work, and some employers face financial incentives not to create jobs.”

That’s why even growing business are now getting by with substantially fewer employees.

When Times Get Tough, the Elderly Work

Copyright, The New York Times Company

The elderly are one group whose work hours now exceed what they were before the recession began. This pattern is most evident in the most depressed regions of the United States.

The recession has varied in different regions of the United States. In some areas – including Arizona, California, Florida, Hawaii, and Nevada – housing prices surged more dramatically in the early part of the 2000s than they did in the rest of America, and their economies fell hard when housing prices collapsed.

One view is that such areas experienced a deeper recession because their banks became overwhelmed with defaults and were unable or unwilling to make new loans to consumers and businesses. Without those new loans, demand collapsed more than it did nationwide, and jobs were especially difficult to find, even while people living in the area were especially eager to work.

Absent demand, just about all workers will have a tough time retaining a job or finding a new one.

Another view is that old loans are the problem, not newer ones. A significant fraction of households and businesses are typically so burdened with the debts they accumulated during the housing surge that they have little incentive to produce and work, because their creditors would get most, if not all, of the fruits of their labor.

In contrast to the no-new-loans-and-no-demand theory, old loans do not affect all workers; some are less burdened by debt. The elderly may fall in this category, because they are more likely to have saved money over their lifetimes and to have paid off their mortgages. Although some elderly working for debt-burdened employers may have lost jobs, on average the elderly in these areas should be working more because they have better incentives to do so.

The chart below compares 2007-10 changes in work hours for two areas –- the regions where housing prices rose and fell the most, on the left side of the chart, and the rest of the United States on the right. For middle-aged and younger people (blue bars), hours worked fell 12 percent in the large cycle regions and about 9 percent in the rest of the United States.

Hours worked by elderly people increased in both regions.

As I noted a few weeks ago, the average American elderly person worked more in 2010 than did the average elderly person before the recession began, even while work hours were down sharply for middle-aged and young people. The chart above shows that this is true even in the states that generally experienced the largest collapse during this recession.

Demand is not the only factor driving employment patterns.

America's Colonial Safety Net

Copyright, The New York Times Company

The United States Coast Guard’s search-and-rescue division is considering some of the same trade-offs that are found in better-known safety-net programs related to unemployment and health care.

Among other duties, the Coast Guard rescues people in, on and near United States waters. In addition to having highly trained life-saving personnel, the Coast Guard has ships, helicopters, planes and some of the world’s most modern life-saving equipment.

Taxpayer financed, the Coast Guard every day offers its services free of charge to the people it rescues. In that regard, the search-and-rescue part of the Coast Guard is a kind of safety net –- taxpayers pay so that people can have help on the rare occasions when they need it.

In fact, these search-and-rescue activities predate Medicaid and unemployment insurance; the Coast Guard can be traced back more than 200 years to a private search-and-rescue organization in Massachusetts called the Humane Society.

Safety-net programs have what economists call “moral hazard” as an unfortunate byproduct: recognizing that the government is standing by to help, some people do too little to take care of themselves. For example, a significant fraction of unemployed people delay finding a new job until their benefits run out. This may not be a choice we condone, but because unemployment insurance makes unemployment a little less painful, some people will respond by doing less to exit unemployment.

In their marine rescue efforts, the Coast Guard has noticed that many boaters do not wear life jackets that can prevent or delay drowning, and a number of boats venture miles offshore without a radio beacon that can help rescuers find troubled boaters.

Unfortunately, some boaters’ imprudent actions may actually be the result of the Coast Guard’s life-saving proficiency. Boaters know that they can call the Coast Guard when their boat takes on water, catches fire or otherwise becomes disabled, and many times help will arrive before the boaters have to leave their vessel and jump in the water. The Coast Guard even has ways of locating people who do not have radio beacons.

President Obama signed a health care law that will require Americans to get health insurance. Although less known, President Obama also authorized the Coast Guard to consider mandating radio beacons for boats venturing offshore and mandating that boaters wear life jacket in many situations (the Coast Guard has long required boaters to have life jackets on board, even if they are not worn).

Although it is sometimes asserted that the federal government has limited legal power to mandate citizens’ purchases, there can be a good economic case for mandates. While economists are often not inclined to interfere when a person knowingly risks his life for thrills or any other reason, the fact is that our government is in the safety net business, so individuals may take risks without recognizing the costs they create for rescuers.

A boat owner may save himself a few dollars by forgoing the purchase of a radio beacon, and that can ultimately cost the Coast Guard –- and thereby American taxpayers –- a great deal of money in search-and-rescue resources. Over all, it might be cheaper if the owner were force to make the purchase.

On the other hand, it is possible (although not logically necessary) that mandating marine safety equipment may encourage dangerous activities on the water, because the safety equipment reduces the costs of dangerous activities. Years ago, Prof. Sam Peltzman of the University of Chicago found that mandating the wearing of seat belts in automobiles resulted in more dangerous driving and more accidents, because, thanks to the seat belts, the average accident was less deadly.

As always, helping people has its side effects, and some rescues are necessary because the safety net exists.

Small is not Zero

Copyright, The New York Times Company

Payroll taxes are by no means the only thing that stops people from working, but one of President Obama’s payroll tax cut proposals could nonetheless create a million or more jobs.

Last week I estimated that the president’s proposal to cut the employer portion of the payroll tax by 3.1 percentage points could raise employment by more than a million, and maybe as much as three million.

You might (as some readers wrote to me) think that a payroll tax cut is not, by itself, a good reason for employers to hire, and on that basis conclude that my estimate is way off.

I agree that jobs are not created by payroll tax cuts alone, and my estimate reflects that fact. About 131 million adults are working now, and 109 million adults are not working. If I’m right that the payroll tax cut would raise employment by one million to three million, that means that 106 million to 108 million adults would still not be working despite the payroll tax cut.

The chart below illustrates the results for the case that the payroll tax cut raises employment by exactly two million.

In other words, my estimate is that at least 97 percent of people not working would still not be working regardless of the payroll tax cut. That’s because, as you might deduce, payroll taxes are only one factor among many that determine how many people are employed. Nevertheless, raising employment by one million to three million would be an accomplishment for the president, and one that would be visible in the national statistics.

By the same logic, if someone were to propose raising the payroll tax by 3.1 percentage points, I would expect employment to be reduced by one million to three million. Again, the payroll tax is only one of many factors affecting hiring decisions, which is why my estimate of a payroll tax increase implies that more than 97 percent of workers would continue to work despite the increase.

Indeed, we all know people who would continue to work even if the payroll tax were raised by 30 percentage points, let alone three. We also know people who would not work even if taxes were eliminated completely.

But the fact that more than 100 million people are not employed, and more than 100 million people are employed, suggests that there could well be a million people (or two million, or three million) who are near the fence. For that small fraction of the population, there are almost as many things pushing toward making them employed as making them unemployed; a payroll tax cut could tip the balance.

For hundreds of millions of others, the balance is tilted too far for a payroll tax cut to make a difference. But while economists can debate the exact numbers, few of us can conclude that a small tax cut has no effect. Rather, a small tax cut should be expected to have a small effect — and at this point one worth seeking.

How Payroll Tax Cuts Can Create Jobs

Copyright, The New York Times Company

Last week President Obama proposed a collection of policy changes, including payroll tax cuts, unemployment benefit extensions and new infrastructure projects. The latter do not have much job-creation potential, because they reduce private-sector activity in the short run. But they can be desirable for the infrastructure they produce, and because doing some of those projects now would be cheaper than doing them later.

Unemployment compensation may be compassionate, and for that reason alone might be the “right thing to do.” But an unfortunate side effect of unemployment compensation is that it reduces employment by discouraging people from seeking and retaining jobs.

The real job-creating potential in the president’s proposals comes from one of its payroll tax cuts.

The payroll tax is the second most important tax in the United States, normally bringing in almost $900 billion a year through a combination of taxes on employers and employees — about 15 percent of payroll. Although workers may not realize it, most of them pay more payroll tax than they pay in federal income tax.

The president proposes cutting the employer portion of the payroll tax by 3.1 percentage points (bringing the combined total down to about 12 percent) for employers with less than $5 million in payroll. Unfortunately, this last condition is business-distorting. Why encourage a $10 million business to split into two $5 million businesses?

Nevertheless, the 3.1-percentage-point part of the president’s proposal could raise employment by at least a million, albeit the duration of job creation is related to how long the tax cut lasts. I expect that every percentage-point reduction in employers’ costs raises employment by about a percentage point and real gross domestic product by about 0.7 percentage point.

That means employment could be roughly three million greater during the period of the tax cut than it would otherwise.

The tax cut is proposed to last a year, and some of the estimated three million incremental job-years — a job that lasts a year, or 12 jobs that last a month — could be spread over time. So we might see only two million in the first year of the cut, with another one million after the cut expires. But still that’s a lot of jobs.

The other part of Mr. Obama’s payroll tax cut proposal is more complicated — and counterproductive. It would reduce the employer’s proportion of the payroll tax by 6.2 percentage points for increases in its payroll spending. Assuming that this payroll tax change would be in place in 2012, the payroll spending subject to the reduction would be the difference between the 2012 payroll and the 2011 payroll.

Because this part of the cut is based on the payroll difference, it makes expanding the 2012 payroll cheaper — presumably the intention of the law — but it makes it cheaper to contract the 2011 payroll.

That could be part of the reason why employment so far in 2011 has been so low; if you think that tax credits for new hires will catch employers completely by surprise, remember that the Obama administration has been floating ideas like this for three years.

To see this, consider an employer that would have a $1 million payroll in both 2011 and 2012. With the normal rates in place, that employer and its employees would owe a combined $150,000 in payroll taxes in each of the two years (15 percent of payroll; for simplicity I have put to the side payroll tax caps and a employee-side cut that has been in place since Jan. 1), or a total of $300,000.

If this employer decided to increase its 2012 payroll by $100,000, that would add a total of about $15,000 to the tax bills under the normal rates, but only about $9,000 under the proposed cut. In other words, as intended, the proposal makes 2012 payroll expansion about 6 percent cheaper than it would be under the normal rates.

However, if the same employer decided to cut its 2011 payroll by $100,000, that would subtract a total of about $15,000 from the tax bills under the normal rates, but subtract a total of $21,000 from the tax bills under the proposed rates — if the payroll cut was restored in 2012, since that part of the payroll would benefit from the reduced payroll tax rate. Contrary to the policy’s intentions, it makes cheaper certain types of payroll reductions, namely those reductions that occur before the law goes into effect.

While President Obama’s proposals have some real job creation potential, it remains to be seen whether any of them become law and whether the job-creating policies are packaged with too many job-destroying policies.

News is What the Customer Wants

A few quick google searches (items published Sep 1 - 7 AM) on coverage of the "Jimmy Hoffa speech."

New York Times 0 search results link to story/discussion
Chicago Tribune 1 search result links to story/discussion
New York Post 3 search results link to story/discussion
Fox News dozens of search results link to story/discussion

Who Lost Work During the Great Recession?

Copyright, The New York Times Company

Young people have seen their work hours drop the most during this recession, while the elderly are actually working more than they did before.

Using data from the Census Bureau’s Household Survey via the National Bureau of Economic Research, I calculated the average hours worked by age for 2007 (people not working during the week of the survey count as zero hours worked) and then again for 2010. The chart below displays each age group’s percentage change from 2007 to 2010. For example, the chart shows that the average 16-year-old in 2010 worked 40 percent fewer hours than the average 16-year-old did in 2007.

Author’s calculation from Census Bureau Household Survey

We all know that hours worked in 2010 were considerably fewer than they were before the recession began, which the chart shows: most of the age groups have a negative percentage change.

But the chart also shows that labor losses lessen with age and are positive for a number of age groups. In percentage terms, work hours fell the most for teenagers, reflecting the high teenage unemployment rate. After the teenagers, work hours fell the most for the age groups 20 to 29. Work-hours losses for groups in their 30s and 40s ranged 5 to 11 percent. Work hours also fell for age groups 50 to 59, but typically less in percentage terms than for the age groups aged less than 50.

As I noted a few weeks ago, average work hours actually increased for the oldest age groups.

Seniority layoff practices would tend to reduce hours worked most for young people because, naturally, they tend to be employers’ more recent hires. You might think it would make sense for employers to retain their most experienced workers, but downsizing employers tend to offer and encourage early retirement to people in their 50s and early 60s, who are paid more than recent hires and are starting to think about leaving the workplace.

Yet the chart does not show especially large declines in hours for those age groups (nor can seniority practices by themselves explain why the elderly end up working more).

Of course, an employer that shuts down does not lay off based on seniority but lays off everyone.

Another possibility is that the labor market distinguishes, at least in a rough way, among workers according to their willingness to work, and that the stock market and housing market crashes have especially stimulated older people to work more. (Young people, on the other hand, had fewer assets before the recession, so a decline in asset prices has little direct impact on them.) This effect tends to increase with age because the propensity to own assets for current needs and future retirement also increases with age.

To the extent that minimum wages reduce employment of people who would otherwise earn a wage less than the minimum, the minimum wage increases of 2007, 2008 and 2009 may be another factor, because propensity to earn near the minimum wage tends to decline with age (although that propensity is not particularly low for the elderly, who do not have work-hours losses on average).

It is also possible that the ability to efficiently find a new job in a tough labor market is a skill, and people tend to accumulate that skill with age.

Economists are still digesting the labor market data from the Great Recession, but for now it appears that getting back to the pre-recession labor market especially requires creating jobs for young people.

Preparing for Disaster

Copyright, The New York Times Company

Hurricane Irene traveled the East Coast last weekend and was expected to be one of the rare hurricanes to hit New York (by the time it reached there, it had been downgraded to a tropical storm). Although many thousands of people in the region may be without electricity for days to come, New Yorkers are glad that Irene did little damage as hurricanes go.

The storm began to receive media attention the weekend before it arrived, and people in the New York area used the time to prepare themselves. Forecasters have been criticized for overestimating the amount of wind that New York would experience, but the winds and tides were plenty strong. Preparation was an important reason that damage was limited.

For example, the eye of the storm passed right over Kennedy International and LaGuardia Airports, but planes did not crash because the airports were shut during the storm. Few lives and little cargo was lost from ships in the area because the ports were closed, and vessels moved out of the area.

Hundreds of thousands of people were left without power, and many homes were flooded. But, clearly, the damage would have been much worse if Hurricane Irene had hit with no warning.

I have no expertise in climate, weather, physics or aerodynamics, but one of the lessons of economics that environmental changes are less harmful when they can be anticipated. With only a few days of preparation, as with Hurricane Irene, people and mobile capital can be moved out of harm’s way. Protective barriers can be constructed for the immobile capital like homes.

New York was not given a definite warning of Irene a year ahead, let alone a decade ahead. But if it had been warned years ahead, the economy could have adjusted even more by making plans to locate activity in more protected places. Or perhaps even to invent goods and production processes that are more hurricane resistant.

The opportunities for preparation are one reason why economists expect the damage from global warming to be different, and perhaps less, than from natural disasters that hit by surprise.

Global warming is expected to raise temperatures and sea levels over a period of decades – perhaps centuries. Even if scientists are completely unable to retard or reverse the environmental consequences of global warming, thanks to decades of warning the economy can greatly adjust to minimize the economic costs of those environmental consequences.

That’s the thesis of a recent book, “Climatopolis: How Our Cities Will Thrive in the Hotter Future,” by Matthew Kahn of the University of California, Los Angeles (disclosure: Professor Kahn is a friend and was my classmate at the University of Chicago). Given advance warning, people will protect themselves and innovate in order to make sales in future market when the symptoms of global warming are more threatening.

A Fallacy of Composition

Copyright, The New York Times Company

In 2009 the New York Yankees opened their new stadium on the north side of East 161st Street, replacing the historic stadium on the south side of the street. Not surprisingly, 2009 spending by consumers, news organizations and entertainment businesses, among others, on the north side of East 161st Street was a lot more than it had been in years past. It all started from the Yankees’ spending at the new location.

So a spending advocate might assert that this episode is proof that spending by one organization can stimulate spending by others, because the spending by the others on the north side of the street surged at exactly the same time that the Yankees started having their people work there.

Of course, such an analysis is flawed, because it ignores what happened on the other side of the street. Much of what happened north of East 161st Street was just a displacement of activity from the south side, rather than a creation of new activity. Even the construction workers building the stadium may well have been drawn from other tasks.

This pattern is not special to the Yankees’ move. A number of studies have shown that consumer spending associated with a sports team to a large degree displaces spending in other areas and displaces spending on other leisure activities; a family is unlikely to conclude that because there’s a new team in town or a new stadium, it should sharply increase its spending on entertainment.

Yet ignoring the displacement effects is exactly what Paul Krugman and Dean Baker have done in their praise of recent studies that use “cross-state variation in stimulus spending per capita to estimate the employment effects of the stimulus,” studies comparing states that received more stimulus to states that received less.

Spending from the American Recovery and Reinvestment Act (a.k.a., the “stimulus”) could be very much like the stadium spending — a locality that received more stimulus spending merely enjoyed a displacement of activity into its area from localities that received less spending, and that nationally little or no additional spending occurred as a result of the legislation.

If you want to know about the national effects of the stimulus, at least part of the analysis has to look at the nation as a whole. The same is true of the national effects of changes in labor supply. If one group suddenly becomes more willing to work, it is possible that the group solely takes jobs from the rest of the population, with no new jobs being created for the nation as a whole.

That is why, in addition to looking at the experiences of specific groups and specific regions, I have examined the effects of supply and demand on national employment. (Professor Krugman and Dr. Baker assert in so many words that I ignore national employment, though my papers on the subject look very much at national aggregates. For example, see Figure 3 and Figure 6 of this paper and Table 2, Table 3, Figure 2, Figure 3, Figure 4 or Figure 5 of this paper).

I found, for example, that national employment increases during the summer precisely because young people are more willing to work. Not surprisingly, the summer surge of young job seekers does seem to reduce employment of the rest of the population, but the net national effect is still almost a million more jobs in the summer.

For now, it appears that government spending reduces private spending, even while it may benefit specific regions or groups.

Exceptions to Keynesian Theory

Copyright, The New York Times Company

While taxpayers have been wondering if all of the extra government spending of the past couple of years has actually served to impede the recovery, Keynesian economists have been asking them to keep faith in the promise that government demand is the secret to economic recovery. Now Paul Krugman, an outspoken Keynesian stimulus advocate, admits that Keynesian theory has many exceptions.

It’s pretty easy to see how various types of government spending might reduce employment, rather than increase it: a number of government programs have been reducing the incentives for people to work, and reducing the incentives for business to hire.

Unemployment insurance is an example (among many) of how the work incentives of so-called stimulus programs operate. Unemployment insurance payments to individuals cease as soon as the individual starts to work again. I agree that such payments are compassionate, and may well be the right thing to do, but economists have long recognized that such compassion is not free: unemployment insurance reduces employment, rather than increasing it, because it penalizes beneficiaries for starting a new job.

Without offering any proof that incentives suddenly ceased mattering, stimulus advocates, and even the Congressional Budget Office, have recently ignored this effect. Many of them aim to prove the potency of unemployment insurance and other components of the stimulus law by insisting that the recession was caused by a lack of demand, and that any public policy that raises aggregate demand must be a big help.

Even if they’re right that the recession was a result of low demand, it does not follow that the way to recovery is to destroy supply, too. Before we turn away from one of the basic lessons of economics, we ought to have some evidence of the fundamental Keynesian proposition that “incentives to seek work are, for now, irrelevant.”

(Another tendency of Keynesians is to “prove” their supply claim by pointing to the existence of unemployment. Of course, unemployment exists in large numbers, but that does not tell us whether, and how much, incentives affect employment rates.)

Part of my research has been to examine episodes, from the current downturn, of changes in the willingness and availability of people to work. If, as Keynesians have been insisting, the incentives to work are in fact irrelevant in a recession, then none of these episodes would be associated with employment changes. (In their view, an increase in the number of people willing to work would just increase, one for one, the number of people who are unemployed.)

I looked at seasonal changes in labor supply. I looked at the increase in supply of workers to the nonresidential construction industry (workers who were leaving home building after the crash). I looked at the increase in the supply of elderly workers. I looked at the increase in supply of workers in Texas. In all of these recession-era episodes, more supply meant more jobs, and less supply meant fewer jobs.

(I also looked at some recession-era demand changes to see if they were at all constrained by supply, and they were — very much as they were before the recession.)

There is still no evidence to confirm the fundamental Keynesian proposition that supply doesn’t matter.

Rather than completely discard that proposition, Professor Krugman has recently formulated a theory of exceptions to the Keynesian theory, which he believes can help explain some of my findings:

Here’s the question: why do patterns of employment over time that are, in fact, normally supply-driven continue to be visible even during a demand-side slump? And here’s the answer: businesses make long-term decisions that influence hiring patterns over time, and those decisions continue to shape their behavior even when there is a surplus of labor.

In other words, Keynesian theory has exceptions that have to do with business’s long-term hiring decisions. For example, businesses have lived through enough seasonal cycles to know that they can normally make more money when their hiring patterns are responsive to the seasonal availability of people to work, so businesses continue to be responsive to the seasonal pattern of labor supply even during a deep recession when there are plenty of workers available throughout the year.

I don’t understand how Professor Krugman explains that the nonresidential construction industry took advantage of the plentiful supply of home builders after housing crashed (he also has no explanation for my minimum wage findings, Christmas seasonal findings or elderly employment findings). He also fails to explains why some business hiring patterns survive the recession intact, while other practices are completely different (e.g., businesses used to think they needed 138 million payroll employees, but by 2009 they got by with fewer than 130 million).

But even if Professor Krugman were correct that the ghost of labor supplies past haunts the recession through business’s long-term decisions, how can he be so sure that the labor-supply effects of government spending programs would not also have the same effects they did in the past?

For example, employers found that people were more difficult to hire and retain when a generous safety net was available. In this way, unemployment insurance would continue to reduce employment even after the recession began because employers have learned that the more generous the safety net, the more they must get by with fewer workers.

Would Keynesian stimulus spending work only when it came as a surprise? Or only when the spending was outside the range of prior business experience? Keynesian economists have not even begun to answer these questions. For now, Keynesian theory has so many exceptions that we might as well discard it.

Supply of Admissions

Professor Krugman is slowly converging to my view that supply matters even during recessions. I'll explain more tomorrow in a more lengthy blog post, but the basic idea has been in my seasonal supply paper (see the final para).

The Debt Downgrade and Stock Prices

Copyright, The New York Times Company

The big financial story in the last few days has been the declaration by Standard & Poor’s that United States government bonds were no longer worthy of its AAA rating. The agency, in a nutshell, thinks there is some chance that the government will default or be delinquent on its debt payments.

The announcement was followed by a wild ride in the stock market in the last two days — a plunge of almost 7 percent in the Standard & Poor’s 500-stock index on Monday, followed by a surge of almost 5 percent on Tuesday.

But with the index down almost 18 percent from its April peak, it is clear that investors’ concerns long predated the downgrade.

The real news is how poorly the economy is doing, and how poor its prospects seem. The chart below shows the changes in several indicators of economic activity over the last nine months. The blue series is an inflation-adjusted stock price index, which (even with Tuesday’s big gain) is lower than it was nine months ago. Through May 2011, real housing prices (black series) were down 7 percent. Real consumer spending (red series) has failed to increase, and inflation-adjusted spending on consumer durables has fallen four months in a row.

All of these indicators are forward-looking in the sense that they depend on what people expect to happen to incomes and profits in the future. These indicators had been looking better during much of 2010 but now it seems that consumers and investors are not optimistic about what is ahead (are they worried about riots like in London? higher taxes? government program cuts?).

In my view, a rating agency does not move the market but rather reacts to some of the same prospects that are reflected in the decisions of consumers and investors. For example, to the degree that incomes continue to remain low, tax collections will also remain low, making it that much more difficult for governments to pay their obligations.

Recovery for the stock market, and the wider economy, needs a lot more than an agency to change its mind about government bond ratings.