What works in fiscal policy.

In the New York Times Greg Mankiw reviews empirical evidence that suggests tax cuts work as fiscal policy, while increases in government spending do not.

The results are striking. Successful stimulus relies almost entirely on cuts in business and income taxes. Failed stimulus relies mostly on increases in government spending.
All these findings suggest that conventional models leave something out. A clue as to what that might be can be found in a 2002 study by Olivier Blanchard and Roberto Perotti. (Mr. Perotti is a professor at Boccini University in Milano, Italy; Mr. Blanchard is now chief economist at the International Monetary Fund.) They report that “both increases in taxes and increases in government spending have a strong negative effect on private investment spending. This effect is difficult to reconcile with Keynesian theory.”

You might almost think that the supply side people were on to something. 

Cash for clunkers

The Cash for Clunkers program that ran out of money after just a few days in operation is an interesting program from many perspectives. It is firmly in the tradition of some of the New Deal Programs--it immediately brings to mind the agricultural program that paid farmers to kill livestock. It seems to have a considerable amount of political appeal because it seems to be effective.

I have not read economists on the program, but I am confident most of them, other than some of the extreme partisans, would give it thumbs down. Economists have certain criteria by which they evaluate programs. One of them is equity or fairness. Who does the program benefit? The problem that this program has from an equity point of view is that it seems largely arbitrary. It is as if the government held a lottery and gave random people several thousand dollars. To qualify, one must have an car that has low value. Lots of lower and middle class families have those, often as a second or third or fourth car. I do not know enough about the rich to know if they tend to have old clunkers. Then one must be willing and able to buy a new car. For some of the poor, that may not be an option--they will not be able to make the payments. As time goes one, someone will figure out if this program subsidized the rich, the poor, or the middle. My guess is that it was a subsidy that went largely to the middle--or to the dealers, who were able to charge higher prices because of the program.

A second criteria is efficiency, which asks the question of whether the program increases value. Here the program is clearly a disaster because it destroys things that have value. The cars traded in with the program must be destroyed, as must their parts. The program is taking a lot of cars that are worth a couple thousand dollars each and converting them into scrap worth a few hundred dollars each. This brings up a secondary equity point. For many of the poor, the purchase of used cars is their best option in getting vehicles. This program will tend to raise the price of older used cars because it reduces their supply. The poor will pay more, and most economists believe that programs that hurt the poor should be condemned on equity grounds.

However, are these bad effects worth enduring for the good of stimulating the economy? We can see that sales of autos have greatly increased as a result of the program--so much so that the program ran out of money after just a few days in operation. A key question here is to what extent were those sales new sales, sales that would not have taken place without the program, and to what extent were those sales simple shifted in time. If you wanted to buy a new car and had a car that qualified as a clunker, you would have had a strong incentive to wait until the program was in force. Or if you were planning to buy a new car sometime in the future, you would have had a strong incentive to move the purchase forward in time to take advantage of the several thousand dollar grant. It is not clear that the program did more than shift sales in time, and if that was its primary effect, it was meaningful stimulus.

Addendum: People find ways to game the system, another example of people responding to incentives.

Update: Econbrowser had a post on the program with many comments.

Fiscal policy as placebo

Robert Frank makes a placebo-effect argument for fiscal policy, something I have not seen before:

John Maynard Keynes once compared investing in the stock market to picking the winner of a beauty contest. In each case, it's not who you think will win, but who you think others will pick. It's the same for those trying to sort out the stimulus debate. The argument of stimulus opponents hinges on their belief that consumers and businesses will predict that stimulus won't work. The fact that stimulus opponents are far less numerous, have less distinguished academic credentials, on average, and are far less ideologically diverse than their counterparts does not guarantee they're wrong. But these factors should make rational consumers and investors less likely to side with them. And since this is really an argument about expectations, that's probably enough.

If we believe it will work, it will work.

(If he is consistent, he should be very harsh on the talking-down of the economy before the stimulus package was passed in February.)

Will the third time be the charm?

There has been buzz about the possibility of a another stimulus package since the first Obama package does not seem to be curing the recession. I was wondering why they were not calling it a third stimulus, but Donald Marron made that point before I could type anything. (The Bush stimulus package of 2008 would be the first one, a stimulus package that was actually properly timed.)

Can we say it failed?

As part of the advocacy for the stimulus plan that the Obama Administration proposed and signed in the first weeks of his presidency, the Obama economists provided a forecast of what the economy would do with and without the stimulus. For a summary, see this graph.

With the unemployment figures for April at 8.9%, it appears that we are not on the track that was projected for what would happen with the stimulus plan, but rather on the track that was projected for what would happen without it. In fact, it appears that we may be above the unemployment rates that were projected if we did not pass the stimulus plan.

If this pattern continues, will be able to say that the stimulus failed? Unfortunately, we probably will never be able to really tell what effect, for good or for bad, the stimulus spending actually had.

An unintended consequence of the stimulus bill

Increased U.S. borrowing to fight the recession will crowd out developing nations from capital markets.

Here’s a paradox for Obama supporters: The first American president with personal roots in the developing world (Kenyan father, Indonesian stepfather, childhood residence in Indonesia) is doing more harm to the developing world than any American in history.

Foreign ownership of the U.S. government debt

The Treasury has a table that shows the top fifteen countries that own U.S. government debt.

An interesting question that some have been asking is what will change with the much increased deficit that the so-called stimulus package will cause. Since the amount borrowed must equal the amount saved, this equation must hold:

(Private savings - private investment) + (Government surplus) + (net lending by foreigners) = 0

Currently the middle term is negative and will become much more negative. The final term is positive--we buy goods from foreign countries, and they use some of the proceeds to lend back to us, buying our debt. Among the things that a fiscal stimulus could do are:
a) Increase private savings--the traditional Keynesian multiplier argument. If fiscal spending increases income, people will save more.
b) Reduce private investment--one of the crowding-out arguments. If the government borrows, it may crowd out private borrowers, reducing investment.
c) Increase net lending by foreigners. One way for this to happen is for the value of the U.S. dollar to rise, reducing our exports and increasing our imports. A decrease in exports provides another way to crowd out the effects of fiscal policy. (A reason the value of the dollar might rise is that U.S. interest rates might rise above the level in other countries, causing a capital inflow.)

Those who argue for a fiscal stimulus package believe the first effect is important relative to the second and third. Those who argue against a fiscal stimulus usually believe that the first of these is unimportant relative to the second and third items.

Time to delete mark-to-market?

There were rumors on Friday that the mark-to-market rules might be relaxed, and those rumors were credited with the rise in bank stocks.

Which leads to the question, why are those rules still in place? There have been suggestions throughout the financial panic that the mark-to-market rules were one of the things that was accentuating the fall. And if we look at how the Federal Reserve has responded, we can see that this financial panic is like no other. What is different this time? Could it be that the mark-to-market rules have introduced amplifying feedback, making the financial crisis a self-feeding process?

When FDR became president, his administration was a lot like the character in a comedy reacting to a crisis by pushing buttons. FDR pushed a lot of wrong buttons, but he did push one correct button by severing the tie to gold and eliminating the Fed from monetary policy, which had established an amplifying feedback loop in monetary policy. Maybe the mark-to-market button is the right one this time. Certainly it is worth trying. If the U.S. is willing to spend nearly two trillion dollars, the amount of the TARP and the so-called stimulus plan, why would it not also be willing to junk a rule that costs virtually nothing at all to junk? I wonder why the Bush administration was never willing to take this simple step given that they were willing other steps that were far more drastic.

Economics lacks a consensus framework in which to view macroeconomics. A framework of shocks and feedback, in which shocks knock the economy off path and the feedback accentuates the effect of the shock, leads one to focus on items like mark-to-market. The equilibrium approach of Keynesian economics leads to a focus on stimulus packages. If we get both a stimulus package and an elimination of mark-to-market, it may be very difficult to distinguish the effects of each.

Update Feb 16, 2009: Brian Wesbury, chief economist at First Trust Portfolios L.P., argues in a National Review piece, "Untouchable Accounting Rules? Really?", that mark-to-market has been a major contributor to the current crisis because of its procyclical feedback properties, and needs to be eliminated. He points to some history:

Fair-value accounting as we know it today is based on rule FAS 157, which was implemented by the FASB in 2007. But it has a longer history than that. Fair-value accounting existed in the 1930s, which was when we had the Great Depression. In 1938, President Roosevelt suspended those rules, and between then and 2007 the economy had no panics or depressions. Maybe its time we put fair value through the shredder once again.

Stimulus and the Great Depression

The Huffington Post has an article about the two views of the Great Depression,

The dominant story in the public mind is of President Franklin Delano Roosevelt's success. It goes like this. Like Obama, FDR comes to power with the American economy haemorrhaging jobs. He believed that, if private industry is withering, the government has to take up the slack by large public spending programmes. He set millions to work preserving green spaces and rebuilding the country's infrastructure.

The other view is attributed to a small number of right-wing economists in the 1980s.

They argued that the American people had been wrong: the New Deal actually made the Depression worse. By borrowing and spending so much, the government created a climate of uncertainty. This made investors hold on to their money - prolonging the despair. It didn't restore private investment, it "crowded it out".

The author, Johann Hari. then argues the first is correct for two reasons. First, when FDR cut back spending and raised taxes in 1936, we had relapse. Second, it was the spending of WWII that finally ended the Depression.

The article skips the interesting questions that are needed to frame the whole debate. We have had recessions every few years for the past two centuries. Most of them have been mild. Nothing else was close to being as severe or as long-lasting as the Great Depression. Why was it so deep? At the time, many thought it revealed an inherent flaw in capitalism, that it showed that Marx was correct. A large part of the intellectual class that grew up in that era accepted this explanation and became Marxist. But when there was no depression after World War II, despite the predictions of people such as Paul Samuelson, an ardent Keynesian, people had to start looking for other explanations. The most convincing explanation that has emerged is a monetary interpretation, arguing that the Great Depression was special because government policy set up an amplifying feedback loop. As things became worse, government (monetary) policy changed to make it even worse, etc. That loop was not broken until FDR neutered the Federal Reserve by abandoning gold. There remains disagreement among economists as to the importance of the real-bills doctrine and the gold standard in contributing to that feedback loop, but most economists who have looked at the episode with any care now see Federal Reserve policy as key to understanding the decline from 1929 to 1933, and also the sharp recession in 1936-7 when the Fed raised reserve requirements. This explanation was developed and popularized by Milton Friedman and Anna Schwarz, but it can be found far earlier, in, for example, a book by Lauchlin Currie.

Compared with other recoveries, the recovery after 1933 was slow. Why was this recovery so slow? One answer, favored by the left, is that because the fall was so deep, normal resiliency of markets was impaired, and government policy was needed. The New Deal was doing the right thing, but just not enough of it. On the other side, the conservative answer focuses not only on expectations and uncertainty (though not primarily caused by borrowing and spending), but also on interference with markets. When a market has a surplus, either a fall in price or an increase in demand will restore market clearing. If there is no increase in demand, then an attempt to keep price from falling will stop that market from finding an equilibrium. In a system of markets, preventing some markets from adjusting can keep the entire system from re-adjusting. The New Deal interfered with price adjustment in many ways--various programs attacked a symptom of recession (deflation), not the cause (insufficient demand). Hence, it is only natural for economists to argue that some New Deal policies made the problems worse rather than better.

The first rule of medicine is, "First, do no harm." It should also be the first rule of economic policy making. Economists have made a convincing case that government policies during the Great Depression did harm. Hari does not seem to take seriously the possibility that the stimulus package being discussed in Washington could do more harm than good.

(If Johann Hari is correct on what got us out of the Depression, shouldn't he be advocating a massive expansion of the military?)

Update: The Wall Street Journal has an piece dated Feb 2, 2006 by economists Harold Cole and Lee Ohanian that emphasizes the negative role of the New Deal's attempts to micromanage markets.

The four effects of fiscal policy

Arnold Kling at the Atlantic business blog looks at the four effects of fiscal stimulus that some economists are debating. The first is the traditional Keynesian multiplier, the idea that additional government spending creates income, inducing the private sector to also spend more. The second is an offsetting effect he calls the household effect. When people are unemployed, they often use their time in some productive way or as leisure, which can also have value. So the additional output comes at some cost in lost household production or leisure. The third he calls the Galbraith effect. How much is extra government spending worth compared to extra private spending? Galbraith thought that extra government spending was vastly more valuable than extra private spending, so he advocated a shift of resources from the private to the public sector. Conservatives usually think that extra government spending is less valuable than extra private spending. The fourth and final effect is the Feldstein effect, and is the distortionary effect of the taxes that will eventually be required to pay for the spending or to finance the deficit.

It is an interesting way to look at the problem because, like cost-benefit analysis, it makes clear what people's assumptions are.

What we know about fiscal policy

Marginal Revolution has some good posts on the recession. One puts the media hype into perspective. Two others suggest that what we know about fiscal stimulus is very limited (rather surprising since introductory textbooks have been teaching it for half a century, and during most of that time there was little mention that we knew very little.)

The proposed Obama stimulus plan

The economic team of the president elect has released a report on their proposed stimulus plan, and links to it have appeared on a number of blogs (for example, here and here). The text talks about how uncertain the estimated effects are:

It should be understood that all of the estimates presented in this memo are subject to significant margins of error. There is the obvious uncertainty that comes from modeling a hypothetical package rather than the final legislation passed by the Congress. But, there is the more fundamental uncertainty that comes with any estimate of the effects of a program. Our estimates of economic relationships and rules of thumb are derived from historical experience and so will not apply exactly in any given episode. Furthermore, the uncertainty is surely higher than normal now because the current recession is unusual both in its fundamental causes and its severity.

However, the graphs give the impression of precision. Here is a key graph from the report, which I lifted from Calculated Risk:



The report assumes that the stimulus plan will total $775 billion of tax cuts and government expenditures (both spending and transfers). The effect these have on GDP depend on the multipliers, which are estimated to total 1.55 for government spending and .98 for tax cuts (after four years). The report than translates the increased GDP into jobs:

We therefore use the relatively conservative rule of thumb that a 1 percent increase in GDP corresponds to an increase in employment of approximately 1 million jobs, or about three-quarters of a percent.

The end result is that they project 3.625 million jobs additional jobs as a result of the program. I am not sure where their no-stimulus baseline is coming from. I scanned the report, and did not see it, but maybe it is in there somewhere.

If we enact the bill, we will get to see what the resulting unemployment series will be. If we do not enact the bill, we will see another resulting unemployment series. However, there is no way to see both series, so we can never really tell how effective or ineffective the bill will be.

I have a lot of skepticism for almost all forecasts even when I respect the forecasters.

Obama on the recession

An AP story reports:

President-elect Barack Obama said Thursday the recession could "linger for years" unless Congress pumps unprecedented sums from Washington into the economy.

That is a very Keynesian view, a belief that the economy lacks strong stabilizing forces. I live in the other camp, the camp that thinks that the economy has strong stabilizing forces over a period of a few years, and the greatest danger is destabilization from the government. What is rather odd is that both views look to the Great Depression for support.

Fiscal stimulus

The National Bureau of Economic Research (NBER) has decided that a recession began in December of 2007. That date means that the fiscal stimulus of the past summer was properly timed--it occurred early in the recession--which raises the question, "If fiscal policy works the way the textbooks say it does, why have things gotten so much worse?"

The current recession seems linked to two major shocks. The primary one is the bursting of the housing bubble and the mess of bad debt left in its wake. The housing bubble itself was encouraged by lending that made sense only by assuming that housing prices would continue to rise, such as zero-percent-down loans, interest-only loans, and the array of sub-prime and near sub-prime lending. Further, large banks and financial institutions thought that they were managing risk, but most of the risk management techniques assumed that the markets would function smoothly. However, when a financial panic hits, markets never function smoothly. Things that work without problems in normal times can fail to function in abnormal times. After episodes such as the stock market crash in October, 1987 and the demise of Long Term Capital Management in 1998, one would think that people running big financial institutions and the regulatory agencies would have been very aware of that problem.

The second and secondary shock was the oil-price run-up, which has now receded.

Some commentators point to the mark-to-market rules not as a source of a financial shock but as an amplifier of the downturn. I do not fully understand their argument, but it seems to be a feedback argument similar to Irving Fisher's debt-deflation theory of depression. A decline in the value of securities causes distress in some financial institutions that have to mark down the value of assets. Because their net worth is affected, they rearrange portfolios trying to flee risk, which causes a further decline in the price of securities.

Fiscal and monetary policy were designed to offset demand shocks, which is what the Keynesian economics that emerged from the Great Depression saw as the source of almost all instability. But is the source of our present recession a demand shock, or is it something else? And if it is something else, why should we expect any stimulus package, no matter how large, to fix the problem?

Addendum: Here is marginalrevolution.com on monetary policy.

A stimulus plan?

Much of the economy is struggling, but gun sales are booming. Maybe we could save the big three auto makers by threatening to ban the purchase of cars in six months, and people will respond by buying now.

Expectations matter.