Friday, October 30, 2009

Must Mortgages be Modified According to Treasury or FDIC formulas?

Mortgages held by Freddie and Fannie must. So do mortgages held by banks taken over by the FDIC. It appears that Citi agreed to follow the FDIC guidelines with its mortgages as part of the terms of its November 2008 bailout.

For those that are left, participation is voluntary. But we cannot conclude that voluntary participation is equivalent to an endorsement of the FDIC guidelines, because servicers may follow those guidelines merely to partake in the significant Treasury subsidy that goes along with following those guidelines. Treasury has budgeted $75 billion for the HAMP program, which it expects to reach 3-4 million mortgages. That's more than $20,000 of subsidy per mortgage!

The table below summarizes reports by Treasury, FDIC, and the Congressional Oversight Panel as to the type and amount of various subsidies. This list is not exhaustive, but I think the largest part of the subsidy is that Treasury pays half of the amount of a borrower's mortgage payment that causes it to be 38% percent of that borrower's income rather than 31%. For example, if borrower annual income was $50K and unmodified annual housing payments were were $20K, the Treasury would in effect be paying $1,750 per year for five years toward the borrower's mortgage.

Mortgage Modification: Mostly by Reduced Interest Payments

I have summarized reports by Treasury, FDIC, and the Congressional Oversight Panel in the Table below. Pooling and service agreements do not permit term extensions beyond a few months, because each mortgage in the pool is supposed to mature at about the same time. Very little of the monthly payment is principal, so there is not much scope for reducing monthly payments by reducing principal. In practice, the principal balance is a actually bit higher after modification because arrearages are added to the previously scheduled principal.

Thus, modifications reduce interest in order to reduce monthly payments to 31% of AGI. As explained in my papers, this creates a marginal income tax rate in excess of 100%.

Thursday, October 29, 2009

Treasury Concerned that Borrowers May "Game" Mortgage Modification Formulas

I agree. The Treasury's only solution has been to make the modification formula secret (see Congressional Oversight Panel, October 2009, p. 47: the secret is about the exact location of the allocation W0 in this diagram).

What about changing the formula to get the implicit marginal income tax rate down below 100%? Or even down to 0%?

White House Admits Overstating Stimulus Jobs

http://apnews.myway.com/article/20091029/D9BKMVMG0.html

Just a "mistake." Am still waiting for mistakes that underestimate the potentcy of the stimulus.

Sweden, Norway, and the Netherlands

I have received a couple of questions about why I only referred to Sweden, Norway, and the Netherlands in my recent post on sick leave.

First of all, I also referred to the "average European" country. More important, I emphasized those three countries because that's where the literature was focused (my post was very clear that it was not original research, but rather a report on what economists had published on the subject). This study noted both the incentives and the outcomes in those three countries (see the abstract shown in its cover page). Several studies have focused on Sweden alone.

One of the academic papers was featured in my chart, and that study very clearly explained "There is a large body of Swedish literature providing empirical evidence of strong moral hazard effects of the insurance system. See, for example, Andre´ n (2001a, 2001b, and 2003); Johansson and Palme (1996 and 2002); Skogman Thoursie; and Henrekson and Persson (2004). Skogman Thoursie (2002), for example, finds a noticeable increase in men’s sickness absence when popular sports events take place." Hence, the title of my post "Home Sick: Another Example Where Incentives Matter."

Apparently, the message that "incentives matter" is unwelcome among most New York Times readers. Customer is king: maybe that's why Professor Krugman always chooses to omit this critical fact from his writings there.

Public Policy Destroys Homes

This August, I released a paper explaining why the mortgage modification programs implemented by the Bush and Obama Administrations were probably increasing foreclosures, rather than reducing them. The paper was based on government explanations of the programs rules, and economic reasoning as to how borrowers and lenders would respond. I said that, absent a federal modification program, lenders would have their own programs that allow more borrowers to be eligible, and thereby more borrowers who would have their mortgage modified rather than being induced to leave their home for the lender.

Recently the Congressional Oversight Panel released a report reviewing the federal modification programs. It noted how some mortgage servicers are included in the program by mandate. Others servicers have the option to participate, although non-participation (ie, modifying mortgages by rules that differ from the federal rules) would cause them to forgo the 20K+ per mortgage subsidy for the entire population of mortgages they might modify. So you can understand why those "voluntary" servicers might follow the federal guidelines even if they thought alternative guidelines made more sense.

Nevertheless, a number of servicers have chosen to forgo the subsidy and follow their own guidelines. When the Congressional Oversight Panel interviewed them, those that responded said "their own modification programs provided borrowers with more aggressive and flexible relief than did HAMP, allowing more borrowers to receive modifications" (COP, p. 66).

I suppose that this answer was filtered through a committee of corporate lawyers, but if there were any truth to it, it looks like economic theory gets it right again!

Where's the GDP Disaster?

Last October, when we were told that spending and incomes were about to collapse, I predicted that "real GDP will not drop below $11 trillion (chained 2000 $)."

Here is a graph of real GDP during this recession, through September 2009.



Admittedly, I had no appreciation last October for the labor market distortions that were emerging (by early November I started to realize that -- see this post), which allowed me to incorrectly predict in October that payroll employment would not drop below 134 million (now it is amount 131 million). But a lot of people (such as the Obama adminstration) overestimated employment and underestimated unemployment, even with the benefit of the data released and public policies implemented after I made my forecast.

I made a number of other forecasts:

  • non-residential investment would be higher during the recession than during the housing boom [turned out to be correct]

  • the housing market would turn around in summer 2009 [turned out to be correct]

  • the bank bailout would not contribute to new lending [turned out to be correct]

  • the government spending multiplier would be less than one (ie, government spending would crowd out private spending) [I think this also turned out to be correct, but, if you wish, we can wait for more data to close the book on this one]



But were the "Keynesian" economists making predictions? How accurate did they turn out to be? Or were their predictions of the sort that can never be verified?

Related to the so-called GDP collapse: Did you know that real consumer spending and real personal income were HIGHER a year after Lehman collapsed?

Residential Investment Increases for the First Time Since 2005

Last December I predicted that the housing market would turn around "summer 2009." A turn around meant prices stabilizing, and construction returning to more normal levels (to the extent that government increased its construction spending, "the fiscal stimulus will add a wrinkle: more of the housing recovery may be in the form of rising prices rather than rising construction activity").

As early as this spring, we saw several housing price indicators confirming this prediction. But the other part of a genuine housing recovery is housing construction, so it was interesting to see the BEA report this morning that real residential investment in 2009 Q3 was higher than in the previous quarter -- the first quarter-to-quarter increase in almost four years.

Productivity Surges in Q3, Again

For almost a year now, I have characterized this recession as "labor falling, productivity rising." This morning's national accounts release for Q3 confirms that the same pattern continues, because real GDP GREW while hours FELL (productivity is the ratio of real GDP to hours).

For those interested in total factor productivity, note that nonresidential investment is fairly low, and therefore not adding significantly to the capital stock. I haven't done the exactly calculation yet, but expect that total factor productivity grew at a (high) rate very similar to average labor productivity.

Wednesday, October 28, 2009

Indicator in Chicago Looking Better

Last spring I suggested how it was a bad sign that the Chicago Bulls were willing to sell me 180 tickets, rather than the usual 90.

Now, with little effort, I have been able to find people who would like to use the extra 90 (and share proportionally in the cost).

Home Sick

Copyright, The New York Times Company

Although the recent health care debate has featured a number of comparisons of Europe and the United States, little has been said about sick leave. Economic research has shown that workers in the Netherlands, Sweden and Norway often stay home sick.

Incentives, and not the flu, seem to be the explanation.

Paul Krugman (among others) has claimed how Europeans are "healthier than Americans by just about every measure." Thus it may come as a surprise that our poor health does not keep us Americans away from work more often than European workers.

A study by the International Monetary Fund showed that American workers were less frequently absent from work for sickness than was the average European (during the years studied, 1995-2003). As shown in the chart below, workers in the Netherlands, Sweden and Norway stayed home sick about twice as much as American workers did.



Economists have been aware of these differences for a while now, and have understood them to be the result of incentives. Quite simply, the financial penalty for work absence in the Netherlands, Sweden and Norway was quite small (as compared to the U.S. and other European countries), and the labor market responded by keeping workers home “sick” more often.

In Norway, for example, the social insurance system may have to pay a sick worker’s entire salary for the duration of a worker’s sickness, and require the employer to provide still further benefits. Under such a system, sick people are less likely to go to work when sick — but healthy people are also more likely to stay home claiming they are sick.

Indeed, a 2004 paper by the Stockholm School of Economics professor Skogman Thoursie found that the Swedish incentives were so strong that a large number of Swedish men reported sick merely to watch sporting events on television.

Thus, none of the studies have concluded that the Dutch, Swedes or Norwegians are sicker than we are. Regardless of whether you think these countries’ sick leave systems are on balance desirable because they allow sick workers to stay home, or counterproductive because they induce healthy workers to feign sickness, the literature concludes that financial incentives are affecting the size of the work force.

Nevertheless, consideration of incentives has itself been mostly absent from recent public policy proposals here in the United States.

From three consecutive federal minimum wage hikes to the marginal income tax rates created by the new home buyer tax credit and proposals to help employees cope with a health insurance mandate, much has happened in the United States to erode labor market incentives. If that’s our future, we’ll find our labor market to be just as sick as the Europeans’ because — whether we like it or not — incentives matter.

Tuesday, October 27, 2009

Home Prices in August

The OFHEO index showed a decline from July to August, which may be notable given that construction costs went up. Yet, over the same time frame, the Case-Shiller index increased even more than construction costs, with the ratio of home price index to residential construction PPI reaching its highest level of the year.

Sean MacLeod also informed me that the RPX index was up a bit in August, although the RPX increase was not enough to offset the increase in construction costs.

A genuine housing recovery will NOT have real housing prices increasing perpetually (or have the huge amount of construction we had 2004-6), but the level of housing prices may need to be a bit higher than it is now in order to sustain normal construction activity. Recall that construction activity depends on the ratio of home prices to construction costs.

Monday, October 26, 2009

Why Did Chicago Hike It's Sales Tax?

The Chicago Tribune has widely publicized the fact that, as of 2008, Chicago had the highest sales tax rate in the country (10.25%), thanks to a vote in early 2008 doubling of the Cook County portion of that tax.

[In case you are not familiar with the city of Chicago, it is located in Cook County. When the Chicago sales tax hike went into effect, Memphis, TN moved into second place with its 9.25% rate.]

What pushed Chicago taxes up? Even as a recession began? Why has it not yet been reversed, even as the recession has obviously deepened?

It's easy to claim, as a number of Tribune editorials did, that the sales tax hike was a bad idea unfortunately and inexplicably hatched by Cook County Board President Todd Stroger, which would have been avoided if some combination of voters, politicians, and bureaucrats had better economic educations.

I agree that the sales tax hike was a bad idea, and believe that the citizens of Cook County would have been better off without it. [I also support mandating University of Chicago educations for Cook County Board members.] But above explanation has a glaring logical flaw: Cook County Board members' economic educations -- however good or bad they may be -- did not suddenly get worse in 2008. For some reason, President Stroger and his supporters once thought that a sales tax rate less than 10% for Chicago was OK.

Although he may not know it, Chicago Tribune correspondent Greg Burns gives a much better answer in his column today.

Like most cities, Chicago has competed for it citizens. Over the years, citizens have come thanks to its amenities -- we love our lakefront -- and have left because of crime, schools, and taxes.

But the nature of that competition dramatically changed since 2006. The Burns article explains how many fewer citizens are leaving Cook County for Will and Dupage counties, in large part because of the immobility created by falling real estate prices (Chicago real estate prices had some of their steepest drops in late 2007) and tough job market.

Did Cook County government sense in early 2008 that their tax base was more captive than usual? And that a captive tax base is ripe for heavy taxation? I think it's a big part of the story.

This story also explains why, despite a noble effort by Cook County Commissioner Tony Peraica and changes of heart by some key politicians, the Cook County sales tax hike has not yet been reversed: the Cook County tax base remains as captive as ever. Real estate prices are lower, and the job market tougher, than when the Board passed the hike. Maybe we should consider ourselves lucky that the Cook County sales tax was not hiked yet again!

That leads me to a prediction: when Cook County residents become mobile again (I guarantee that will happen eventually), Chicago will cease to be the highest sales tax locale in America. And likely Chicago will relinquish its tax leader status via a rollback of the infamous Cook County Hike of 2008.

Wednesday, October 21, 2009

Stimulus Accountability

as Professor Mankiw has said, there is nothing that can happen that would cause stimulus advocates to confirm that the stimulus law did not create jobs. Even, for example, if employment gains were promised for all 50 states, yet only 1 state saw employment rise after the stimulus.

The Panic of '08: Recession Cause or Effect?

Copyright, The New York Times Company

The financial panics of last September and October will always be part of the story of this recession, just as bank failures are always part of the Great Depression story. But recent research questions the claim that the financial panics themselves contributed to their contemporaneous and severe employment downturns.

In his academic research, Ben S. Bernanke blamed part of the Great Depression of the 1930s on banking panics. And this time last year (at the height of the panic in the commercial-paper market) he was telling President Bush that if “we don’t act boldly, Mr. President, we could be in a depression greater than the Great Depression.” A lot of taxpayer money was spent based on this theory.

Some recent research supports an alternative view: that those financial panics did not cause depressions, but are merely symptoms of deeper economic forces.

The U.C.L.A. economics professor Lee Ohanian’s recent paper has looked at monthly data from the 1930s and finds that bank failures came well after manufacturing establishments had sharply dropped their work hours. Moreover, the banks failing during the initial panics were known to be weak. Whatever brought those weak 1930s banks down had already hit the manufacturing sector hard.

The timing was different in this recession — the largest employment drops seemed to come immediately after the financial panic — but a recent paper by Ravi Jagannathan, Mudit Kapoor and Ernst Schaumburg of Northwestern argues that the coincidence is just as misleading. They argue that the changing global economy — with more employment of residents in developing countries like China — created a glut of savings in those countries, and was destined to reduce employment in developed countries regardless of whether there had been a financial panic.

The foreclosure crisis is not fully behind us, and the time may come again when it looks like “banks are in trouble.” When that time comes, will taxpayers still believe Mr. Bernanke’s theory that they are better off financing bailouts than letting a bank panic run its course?

Thursday, October 15, 2009

Real Hourly Earnings Rise 4.4 percent for the Year

The BLS reported that real average hourly earnings in Sept 2009 were 4.4 percent higher than they were in Sept 2008.

If the combination of productivity and capital had not increased, it would take a 15 percent employment decline to make employers willing to pay so much more.

In fact, employment and hours have not fallen anywhere close to that amount. That's because producivity has surged, which tells us a lot about how this recession's economy is different from that of previous recessions.

Wednesday, October 14, 2009

Another Entry for the Employment-Reducing Policies List

The list of employment-reducing public policies grows yet again:


  • mandating the employers with large payrolls provide health insurance, but that employers with small payrolls do not,
  • means-tested mortgage modification (presenting millions of workers with implicit tax rates in excess of 100% (sic)),
  • means-tested new home buyer credit,
  • mean-tested student loan modification,
  • unemployment insurance extensions,
  • state income tax hikes,
  • IRS means-tested enforcement of prior tax debts,
  • marginal federal tax rate hikes on the "rich"!

Military Recruiting Success: What Does it Say About National Employment?

The military has enjoyed recruiting success lately, despite the rather vivid risks associated with "travel" to Afghanistan and Iraq.

I see several interpretations of this observation:

  1. private spending has fallen, and is an important determinant of total labor demand. Lots of people are looking for a public sector job, and the military is a natural public sector destination for young people.
  2. the composition of private spending has changed from consumption to exports, and young people have a comparative advantage in retail activities. Thus, their effective private sector productivity has fallen (even while productivity has risen more generally), and they are shifting to the military
  3. the federal minimum wage is especially binding for young people that might work in the private sector. Even if we ignore the fact that the law dictating that minimum wage does not apply to military salaries, it certainly applies only to PECUNIARY pay. For example, if an employer paid young people $20 per hour in money, but exposed them to $19 per hour in bad-working conditions (safety risk, time away from family, early morning bugle calls, etc.), the net wage of $1 per hour is perfectly consistent with the federal minimum because it applies to the $20. Thus, even if aggregate labor supply shifted in, the military would enjoy a relative employment advantage because of the composition of its compensation.
  4. some other labor market friction (search, perverse employer incentives, student loan modification with special provisions for public sector employees, etc.) does more harm to private sector employment than to military employment. Thus, as in (2) above, the military is seeing its part of the reallocation of labor from private to public sector.

Obviously, some who call themselves "Keynesian" would stop at (1), despite the lack of a coherent explanation for their spending-employment hypothesis that meets the facts about supply shifts during this recession. I tend to think it is a combination of (2) and (4), but in any case the causes of this recession may be most visible in the market for young employees.

It's Smart to Delay Hiring Early in a Recession

Copyright, The New York Times Company

Politicians in Washington are now considering a tax credit for businesses that expand the size of their work forces. Unfortunately, the mere fact that they are considering this policy is doing harm even before it becomes law.

One problem with the credit comes from the basic arithmetic of employment. The number of an employer’s new hires during a year is equal to its employment at the end of the year minus its employment at the beginning of the year plus the number of people laid off (or quitting) during the year.

Although a credit for new hires could be obtained by having larger payrolls at the end of the year, it could also be obtained — without any change in the size of the payroll – by laying off more people during the year and then replacing (or rehiring) them at the end of the year if and when the subsidy kicks in.

Lawmakers may anticipate some of this behavior by basing the credit on the year’s net employment change: effectively netting out separations (quits and layoffs) from new hires. But of course, this year is not the only year when layoffs occurred; another way a business could enhance its new-hires tax credit is to lay off more employees last year, or at least delay last year’s hiring until this year.

In this case, anticipation of a new hires tax credit was actually making the economy worse before it made it better.

You might say that employers last year had no inkling that a new-hires tax credit was on the horizon. And they can’t exactly go back in time and increase last year’s layoffs to set the stage for a boost this year.

But I expect that a number of employers were aware that government help, when it comes, would be more available for those businesses “in the most trouble.” What better way to spotlight trouble with your business than to lay people off? That’s what happened with federal and state tax policies in some previous recessions. Even this comic strip illustrates the common knowledge that more help goes to businesses that appear to be in greater trouble.

Recall from a previous post a parallel situation with Chicago taxis.

Chicago taxi owners have been delaying their purchases of hybrid taxis because they anticipate that the Chicago City Council will someday pass a subsidy for purchases of hybrid taxis. I doubt that Chicago taxi owners are all that distinct from other American employers in their propensity to think ahead about how next year’s public policies will affect them, and to adjust the timing of their activities accordingly.

Suppose that hardly any employer expected that a new-hires tax credit would be considered before this recession was over. Then it seems that a new-hires tax credit genuinely could do a lot to raise employment.

But how democratic is our government when the success of its policies depends on tricking its own people into thinking it won’t do what it plans to do?
-------
One of the most famous papers in macroeconomics is Robert Lucas' 1976 "Econometric Policy Evaluation: A Critique". There Lucas explains how investment tax credits can be destabilizing because investments are delayed until the government comes along (late in the recession) to offer the credit. I am making the same argument above, because "investment" is nothing more than "new hires" of capital.

Monday, October 12, 2009

Oil is not Food

Professors Becker and Posner say that SUPPLY is a fundamental part of the analysis of food and oil markets.

Friday, October 9, 2009

Another (40 Percent) Marginal Tax Rate

I have emphasized the large implicit marginal tax rates faced by homeowners (with mortgages under water), but renters face a large implicit marginal tax rate too. Professor Kevin M. Murphy writes me:

"You might want to note that the $8,000 new home buyer tax credit provides an additional 40% marginal tax for single households with between 75K and 95K in income and married households with between $150k and $170K of income."

If you are single and a first time home buyer (thus, single but NOT currently a homeowner), you get the full $8000 if you earn less than $75K but nothing if you earn more than $95K. Thus, by earning $20K beyond $75K this person gives up $8 in tax credits (not to mention paying the usual income and payroll taxes). 40% = 8K/20K.

The list of employment-reducing public policies keeps growing:

  • mandating the employers with large payrolls provide health insurance, but that employers with small payrolls do not
  • means-tested mortgage modification (presenting millions of workers with implicit tax rates in excess of 100% (sic))
  • means-tested new home buyer credit
  • mean-tested student loan modification
  • unemployment insurance extensions
  • state income tax hikes,
  • IRS means-tested enforcement of prior tax debts
  • marginal federal tax rate hikes on the "rich"!


Wednesday, October 7, 2009

Did the TARP Increase Lending?

Copyright, The New York Times Company

Economic theory casts significant doubt on the claim that public purchases of bank equity would cause banks to lend more. Now the government’s own watchdog confirms the theory.

During last year’s financial crisis, regulators and market participants grew alarmed at the low levels of bank capital. This motivated the bank bailout, and promises to the public that the bailouts would get the banks — especially nine “healthy banks” targeted by Treasury officials — lending again.

Bank capital refers to the excess value of banks’ assets over their liabilities. Bank capital belongs to the bank shareholders, but provides a degree of insurance to the bank’s creditors — its depositors and bond holders — because their claims on bank assets are senior to those of bank shareholders. Some claim that adequate bank capital is also essential for lending.

The Federal Reserve and the Bush administration let some banks fail last year, but ultimately desired to do something to replenish bank capital. They convinced Congress that they could do so, and had $700 billion (almost $7,000 for every United States household) earmarked for that purpose. Almost $300 billion of that amount had been awarded to banks between late October 2008 and inauguration day, in the form of Treasury purchases of newly issued bank stock.

Officials never admitted to the taxpayers (whose money they requested) that the marketplace might largely, if not entirely, thwart their recapitalization efforts. The market might well react to Treasury share purchases by reducing private holdings of bank capital.

Nor did officials admit that, even if the bailout helped replenish bank capital, banks might not want to use their newfound capital for lending. This was also a relevant consideration, because it is possible that lending opportunities determine bank capital, rather than the reverse.

Part of the bank bailout law established an Office of the Special Inspector General for the Troubled Asset Relief Program (a.k.a. “SIGTARP”) “to conduct, supervise and coordinate audits and investigations of the purchase, management and sale of assets under the TARP.” The inspector general was appointed by President George W. Bush and approved by the Senate.

With the better part of a year to examine the evidence, SIGTARP released an audit report on Monday. It concluded (p. 30) that “…lending at [the nine targeted banks] did not in fact increase….”

The report goes on to lament that this episode could “damage the trust that the American people have in their Government.”

I would put it stronger: This episode is an expensive example of public policy promises that were doomed to failure because they were known at the outset to defy economic theory.

Friday, October 2, 2009

Women's Share Back to 49.8%, BUT

Women's share of payroll employment (seasonally adjusted) fell back to 49.8% in August. HOWEVER, total employment fell to 131 million in September, which is what I said was needed for a female share over 50%.

Gender breakdown for Sept not out yet.

Thursday, October 1, 2009

Professor Delong Sets the Table for Multiplier Measurement

Professor DeLong explains how government spending should have essentially the same multiplier as housing construction spending:

"Shocks to spending boost nominal demand: whenever any significant group decides to boost their spending nominal demand rises, whether that group is new businesses seeking to profit from technological progress in high-tech in the 1990s, construction companies tht find they can obtain cheap financing via derivatives in 2004, or the U.S. government in 2009. The government's money is as good as anyone's." [emphasis added]

If he's right, then we could look at the effects of housing construction spending to learn about the controversial government spending multiplier. Furthermore, Professor DeLong explained to us (above) how the housing boom and bust are especially interesting autonomous spending changes (that is, changes well suited for measuring multipliers). So let's look:



During the boom, I see an obvious crowding out (as opposed to the "crowding in" claimed by Professor Krugman): more housing construction was reducing non-residential construction, not increasing it. Remember that the multiplier advocates claim that spending in one sector motivates spending in another.

During the early part of the bust -- though the end of 2008, and thus through 12 months of recession -- I also see a clear crowding out: less housing construction was increasing non-residential construction. That non-residential construction spending increase did not need a spending boom in the housing or any other sector to get going (Krugman's view) but rather needed LESS spending in other areas.

Even during 2009, it's difficult to see housing spending drops doing much to significantly pull down non-residential spending. To the degree that non-residential spending fell, it may be due to crowding out by public construction spending (not shown in the graph) ... when I get some numbers on this I'll let you know.

Construction Spending Rises

Here's an update to a chart I have shown regularly. Housing PRICES showed a bottom a while ago. Due to crowding out from the fiscal stimulus, it was expected that a housing recovery would be more visible in housing prices than in housing CONSTRUCTION. Nevertheless, it would not be a housing recovery unless construction increased to more normal levels, which is why I have watched the construction activity closely.



I have expected non-residential construction to be high, and it is compared to most of 2007 and all of the years before that. But it appears that construction spending now is just a bit lower than it was when the recession began.

Note that this graph is construction SPENDING, which differ from the VOLUME of construction according to construction prices. 2009's construction prices must be especially low, so a time series for the volume of construction activity should show 2009 to be less below prior years' than shown by the spending series.

More Evidence on Crowding Out

See Professor Barro's recent WSJ, or the academic version.