The Free Thinker

Publicity for "Economic Contractions in the United States: A Failure of Government"

From my co-author Charles Rowley, in the Daily Telegraph: "Adam Smith would not be optimistic in today's economic world: Adam Smith once commented that "there is a great deal of ruin in a nation". He meant that bungling governments imposed only a limited check on the economic performance of a Great Nation."  This article accompanies the launch of Economic Contractions in the United States.  For regular readers of this blog I'll stop to point out my... differences of emphasis... as appropriate:

As Nathanael Smith and I show in our study of US economic contractions, Adam Smith would be much less sanguine were he confronted by today's financial crisis and the US government's response. Indeed, it is not impossible that the US will experience the kind of economic collapse from first- to third-world status experienced by Argentina under the national socialist governance of Juan Peron.

"Not impossible" is so weak that it makes the claim innocuous, I suppose, but this seems like an exaggeration, and it seems to conflict with the 7th chapter of Economic Contractions, where we predict that "profligate deficit-spending and interventionism" will produce disillusionment and a conservative revival (p. 97)-- a prediction that events seem to be bearing out.

The US economy suffers from a growing culture of indebtedness that has increasingly contaminated the federal government since 2001 and has spilled over dramatically into private household behaviour. The combination of the ill-conceived fiscal-furnace fired by President Bush and the US Congress and the reckless monetary-furnace fired by Alan Greenspan and Ben Bernanke throughout the period 2001-2007, created unsustainable housing market and stock market bubbles whose collapse brought on the financial crisis and economic contraction of 2008-2009.

I would put the charge in more measured terms.  Fiscal policy under George W. Bush was less expansionary (smaller deficits as a share of GDP) than under Ronald Reagan, even though the US was at war in the Bush years and not in the Reagan years.  (At a time when Obama and the Democratic Congress are redefining "reckless" in the upward direction at a rate of $100 billion a month or so, it seems strange to use the term for the Bush administration, though the Bush administration should certainly have known better.)   And while interest rates stayed low in 2003-04 for too long after recovery had begun, they then rose steeply, so "throughout the period 2001-2007" seems to make the charge too broad.  Even the low interest rates of 2003-04 seem more like a serious technical mistake combined with having the wrong Fed policy rules-- housing should have been included in the CPI-- than "reckless."  And dating the problems to 2003-04 would seem to absolve Bernanke and concentrate responsibility on Greenspan. 

What led to the bubble was that (a) fiscal and monetary policy were expansionary at the same time for the first time since the 1970s, and (b) house prices were-- we argue, mistakenly-- excluded from the price index, causing the Fed to see "low" inflation at a time when house prices were rising at well over 5% per year.  What I find ironic is that, in a sense, the crisis is Osama bin Laden's revenge.  Bin Laden wanted to deal a devastating blow to the US economy.  Determined to deprive him of this victory, Bush and Greenspan adopted expansionary fiscal and monetary policies that lifted the economy out of what could, and maybe should, have been a deep recession in 2001.  As a result, they inflated a debt bubble and made the next crisis sooner and deeper.  Fortunately, al Qaeda was ruined by the war in Iraq, so bin Laden's victory is Pyrrhic.  But if the government had let the 2001 recession play itself out naturally, there might not have been any 2008 crisis.

The policy responses to the debt bubble demonstrate crude political consideration rather than economic understanding. If excessive government indebtedness is a major source of the problem, why increase the government debt? Why encourage households to go yet further into debt?

The prognosis is catastrophic if projected government policies are not cut back. According to the White House's own estimates, the federal budget deficit in 2009 will be $1.6 trillion, approximately 11.2pc of the overall economy, the highest on record since the end of the Second World War. In 2019, the national debt will represent 76.5pc of the US national economy, the highest proportion since just after the Second World War. In such circumstances, the international reserve status of the US dollar will not survive. As it fades, so interest rates on government securities will rise and the real burden of servicing the debt will increase. In such circumstances, the US economy will teeter on the edge of a black hole.

Hmm... why bring in "the international reserve status of the US dollar?"  I don't remember anything about that in the monograph.  Do we have-- does anyone have-- a theory relating government debt to the viability of a reserve currency?  Personally, if there were some way to transition out of a dollar regime without global chaos, I think it might be nice not to own the world's reserve currency: I think being able to print the world's reserve currency causes a form of "Dutch disease," undermining the competitiveness of our industries and fueling excessive growth of government.  Also, why would more debt cause the US economy to "teeter on the edge of a black hole," rather than, say, acting as a drag on growth and depressing living standards for decades-- as it has arguably done in Japan?  Government debt in Japan and Italy is way above 76.5% of GDP; both countries have suffered malaise rather than crisis.

Prosperity and full employment in the US will only be restored by a return to laissez-faire capitalism. Our study outlines a radical, but politically feasible, approach. Monetary policy should be expansionary. But, on the micro-economic side, tariffs and other trade barriers should be repealed unilaterally; a "Right-to-Work" Act should reduce the minimum wage and curtail the powers of unions; and business regulation should be reduced. Individual banks and their counterparties should not be bailed out, although the system should be protected by ensuring that failing banks are wound up in an orderly fashion – this is the only way to restore market discipline.

A little late not to bail out individual banks, no?  Rather say that laws and policies should be rewritten so as to prevent future bailouts (except of insured deposits through the FDIC).  And is it really "politically feasible" for "tariffs and other trade barriers [to] be repealed unilaterally?"  Maybe!  I wouldn't rule out the possibility that bold, visionary political leadership can make the politically impossible possible -- after all, Obama seems to have made people dislike unions and appreciate the health care status quo! -- but that is as much as to say that the range of the "politically feasible" is very broad.

Rowley and I found enough to agree on to put a book together, but he might have a tendency to overclaim. 

MORE: "Barack Obama accused of making 'Depression' mistakes: Barack Obama is committing the same mistakes made by policymakers during the Great Depression, according to a new study endorsed by Nobel laureate James Buchanan." (Daily Telegraph):

"Endorsed by Nobel laureate James Buchanan."  That's brilliant marketing.  But it's not false advertising, for here is Buchanan's Frontispiece:

A comprehensive and coherent assessment of the current economic contraction, one that largely rebuts attribution of failure to capitalism or the market. The targets become practicing politicians of all parties, whose cumulative mistakes have now hurt us all. We have learned some things to reduce the severity of the current contraction. But we have made no progress toward putting limits on political leaders, who act out their natural proclivities without any basic understand of what makes capitalism work.

Anyway, here's what Edmund Conway of the Telegraph has to say:

His policies even have the potential to consign the US to a similar fate as Argentina, which suffered a painful and humiliating slide from first to Third World status last century, the paper says.

There are "troubling similarities" between the US President's actions since taking office and those which in the 1930s sent the US and much of the world spiralling into the worst economic collapse in recorded history, says the new pamphlet, published by the Institute of Economic Affairs.

In particular, the authors, economists Charles Rowley of George Mason University and Nathanael Smith of the Locke Institute, claim that the White House's plans to pour hundreds of billions of dollars of cash into the economy will undermine it in the long run. They say that by employing deficit spending and increased state intervention President Obama will ultimately hamper the long-term growth potential of the US economy and may risk delaying full economic recovery by several years.

Excellent.  The title of the Telegraph piece may slightly overplay the 'Depression' analogy -- while we do identify some similarities with policies from the Depression era, monetary policy is (no thanks to Obama) just the opposite, and we haven't seen anything like the crazy tax hikes of Hoover and FDR yet though there are some ominous signs -- but this last paragraph nicely captures a central argument of the book.  Conway continues:

The study represents a challenge to the widely held view that Keynesian fiscal policies helped the US recover from the Depression which started in the early 1930s. The authors say: "[Franklin D Roosevelt's] interventionist policies and draconian tax increases delayed full economic recovery by several years by exacerbating a climate of pessimistic expectations that drove down private capital formation and household consumption to unprecedented lows."

Heh.  If Rowley and I get credit for debunking the myth that FDR's Keynesian New Deal got us out of the Great Depression, that would be as unfair as Columbus getting credit for debunking the myth that the world is flat.  To our cost, this myth remains potent at the popular level, but among economists I don't think even Paul Krugman would defend it.  Rather, Keynesians claim that WWII got us out of the Great Depression.  This is also debatable -- "GDP" grew, but should bombs and guns and tanks count as GDP, when they don't contribute to human happiness; or if they do count as GDP, don't we need some more appropriate way of measuring the economy's contribution to human well-being, which would likely put paid to the odd notion that WWII was an economic golden age? -- but it has no bearing on the episode Rowley and I deal with, namely the 1930s.  It's rather odd that at this late date anyone can make headlines with what ought to be a truism, namely that the New Deal exacerbated, rather than "got us out of," the Great Depression, but if that's still news to a lot of people, then those headlines need to be made.  Conway continues:

Although the authors support the Federal Reserve's moves to slash interest rates to just above zero and embark on quantitative easing, pumping cash directly into the system, they warn that greater intervention could set the US back further...

And then proceeds with quotes from Rowley and Buchanan, already quoted above in this post.  I'm glad that the message is clear that we support quantitative easing.  On that point we're basically Milton Friedman-type monetarists.  Too bad our advocacy of immigration reform and Social Security reform doesn't seem to be making it into the reviews so far!

How Keynesianism Survives

"Why the Stimulus is Helping the Economy, But Not Obama" (Time)

Proving a negative is always a challenge, but there's mounting evidence that the controversial $787 billion stimulus bill is achieving one of its major goals: shortening the recession. Economists at Goldman Sachs say that the bill, officially called the American Recovery and Reconstruction Act, has resulted in a 2%-3% boost to annual GDP in the second and third quarters of this year, turning what could have been a worsening recession into potential growth. For President Barack Obama, whose poll numbers have dropped precipitously from around 65% to around 50% as Americans have become worried about government spending and health-care reform, that should be good news: he fought hard against Republican opposition to push the bill through Congress as one of his first legislative acts, and the fact that it's working should be a vindication.

But it's not that easy. Recent polls show widespread disapproval of the stimulus bill: 51% of Americans polled last month by USA Today and Gallup said they thought the government should be spending less under the plan, while 44% said they thought the government was spending the right amount, or should be spending more. If anything, it seems, the stimulus plan is hurting the President even as it helps the country.

The most obvious reason for that disconnect is jobs: despite the signs of a turnaround, unemployment remains stubbornly high at 9.7%, with employers cutting 216,000 jobs in August. While jobs always trail economic rebounds, the unemployment number is higher than economists thought it would be, even in the worst case scenario forecast by the Treasury department's "stress tests" last spring. The point is not lost on Republicans, some of whom have argued lately that no more of the stimulus money should even be spent. "The metric of this bill was job creation," says Don Stewart, spokesman for the Senate's top Republican, Mitch McConnell, "and it hasn't done that."

Still, one would think that the mere infusion in six months of $88 billion into the economy — that's how much the government has spent so far...

Journalists, it goes without saying, are not very smart (maybe they suppress it because it impedes their ability to slice a complex world into neat stories) but a cruder, more flagrant example of the post hoc ergo propter hoc fallacy would be hard to find than the above.  It seems quixotic, or Sisyphean, to remind people of the truism of economic history that it's normal for the economy to recover, that it always does so, even-- nay, especially-- when the government does nothing.  The last time the US government did nothing in a recovery-- in the recession of 1920-21-- the recovery was especially robust.  Again, in 1997-8, when the IMF forced East Asia to adopt austerity measures in the midst of recession, it earned itself the contempt of East Asians but rapidly restored growth to its former booming rates for another decade.  By contrast, Keynesian policies failed in the US in the 1930s, in Japan in the 1990s, and dragged post-WWII Britain down from being the world's economic leader to a desperate, stagflation-ridden basketcase in the generation after the British economics profession surrendered to Keynes, before it was rescued by Thatcher.

So how does Keynesianism survive?  Because of articles like the above, because of journalists committing the post hoc ergo propter hoc fallacy, while economists like Brad DeLong and Paul Krugman, who want bigger government because their values are different from the mainstream and want to use their technocratic credentials to sell big government to people who would not support it if they understood its true effects, back them up.  Keynesian economics is like fortune-telling: shrewdly vague, it exploits a perennial fallacy, namely people's chronic inability to understand that the economy can recover on its own.

In this case, the journalist provides the key fact that makes the "stimulus worked" case implausible: only $88 billion has been spent, just over 10% of the whole stimulus package, and only about 0.6% of GDP, and all of that after the worst of the crisis-- October 2008-- had already been mitigated.  By contrast, Ben Bernanke's Fed has bought hundreds of billions of dollars of private and federal debt, and the Bush administration "spent"-- albeit with a fair prospect of recovering-- $700 billion or so through the TARP program.  If any stimulus is working, it's that of Bernanke and Paulson, which was not mere Keynesian spending-for-spending's-sake but a timely intervention by highly competent technocrats (we won't see such expert competence from neophyte Obama, but fortunately, we might not need it) targeted to the specific problems at the heart of systemic failure.  That's not to give unqualified approval to Paulson.  He's a tragic hybrid of hero and anti-hero: he probably did save the US from a more shattering crisis, but the fallout from the Paulson Plan is likely to be a more politicized financial sector and hugely increased incentives and opportunities for rent-seeking for years or decades to come.  There's something to be said for the House GOP who voted against him, too.

For now, the American people seem to be displaying a sensible incredulity about the "stimulus is working" message.  But don't count on it to last.

"GDP: Nothing to write home about" (Curious Capitalist):

Still, there were some interesting numbers within the GDP report. Consumer spending was down 1.2%, after rising slightly in the first quarter. Private investment was down 20.4%—not nearly as bad as the first quarter's spectacular 50.5% decline, but still pretty danged bad. What kept the GDP contraction to just 1% was a 5.6% rise in government spending (made up mostly of a 10.9% increase in federal spending) and an improvement in the balance of trade: imports fell 15.1% while exports fell only 7%.

What's it all mean? Conference Board chief economist Bart van Ark says the takeaway is that "with capital spending still falling and unemployment rising, neither investors nor workers are likely to see strong rewards anytime soon."

This is why the stimulus was a bad idea. Government spending crowds out private investment, which is the real source of long-run growth. Old-fashioned Keynesian theory says the government should spend to revive the economy in recessions, but old-fashioned Keynesian theory isn't worth much generally, and it's dead wrong right now. The government needs to get out of the way so the economy can grow.

Savings Responds to Incentives

"Americans Get Poorer More Slowly" (Barron's)

It's no news that the plunge in home prices continued unabated. But in past cycles, when stocks suffered their usual declines, residential real estate would only increase at a slower rate -- a fall in the now-familiar second derivative, or the inverse of the current "green shoots" phenomenon. 

Indeed, household wealth is down $13.8 trillion from its peak in the second quarter of 2007, as Michael Feroli of JP Morgan Economics observes. But if the stock market manages to hold its gains through the June 30 -- and the Wilshire 5000 is up $3.4 trillion since March 9 -- this quarter could show the first increase in eight. 

As of the end of the second quarter, the wealth bubble that began to inflate in early 1990s had been deflated. The ratio of household net worth to personal disposable fell to 4.67 times, the lowest since the third quarter of 1992, near the end of the first George Bush's term. 

Back then, Feroli points out, the personal savings rate was 7.3%. In the latest quarter, the savings rate recovered to 5.7%. Savings actually had turned negative as Americans bought into the delusion they could borrow their way to wealth. Why save if you could treat your house as an ATM? 

As the bubble burst, households were hunkering down, paying off debt, boosting savings and in the process sharply increasing their purchases of Treasury securities.

This seems to be consistent with the following story: People want to smooth consumption over the life-cycle, and they seek to achieve this by trying to ensure that their net worth follows an upward trajectory.  If their net worth is rising because of rising asset values, they won't save much.  If asset values are stagnant or falling, they'll save.  

My guess is that it takes people a little while to figure out ways to adjust their spending (e.g., plan cheaper vacations, learn recipes that will reduce their food bills, move to apartments with lower rent or forego moves to more expensive places, etc.), so savings rates will keep rising even if there are no new shocks, though of course at some point they'd come back down because people will have brought their net worth up to the desired trajectory under conditions of lower expectations.

Anyway, what this shows is that savings is not an automatic, fixed share of income, as in the Keynesian model, but is a macroeconomic variable that policy should take into account, and to some extent try to predict and manipulate.  Savings behavior (including failure to save) is rational.  What is needed now is to unclog the channels from savings to investment by creating an environment where businessmen do not fear fresh shocks from policy.  But Keynesian activism prescribes the opposite.  People's desire to save is being accommodated (and to spare) by runaway deficits, but that channels the money into (probably) government waste instead of productive investments that will raise future income.

Martin Wolf:

Economists disagree fiercely on these points [i.e., the need for and effectiveness of fiscal policy]. My approach is “Keynesian”: in extreme moments, the excess of desired savings over investment soars. Again, monetary policy, while important, becomes less effective when interest rates are zero. It is then wise to wear both monetary belt and fiscal braces.

A deep recession proves there is a huge rise in excess desired savings at full employment, as Prof Krugman argues. At present, therefore, fiscal deficits are not crowding the private sector out. They are crowding it in, instead, by supporting demand, which sustains jobs and profits.

Note the strangeness of the Keynesian language.  What could "desired savings" mean?  Surely we would all like to have more savings, ceteris paribus!  Keynes and Keynesians mean, I think, desired savings given incomes, but for savings to be a function of incomes only, and not of interest rates, is inconsistent with rationality.  If the interest rate rises, current consumption becomes expensive relative to future consumption, and I should wish to save more; and the reverse if the interest rate falls.  The opposite holds with investments: the lower the interest rate, the larger the number of investments whose prospective risk-adjusted returns are greater than the prevailing interest rate.  So, if desired savings exceeds desired investment, that means interest rates are too high.  Interest rates should then fall and equilibrate the capital market. 

At the heart of Keynesianism is a simple refusal to believe in this logic.  It is, admittedly, rather hard to prove empirically that interest rates adjust to equilibrate capital markets, given that they are so complex and noisy, and so integrated that there is in effect only one data series (one world capital market) to study.  It is hard to disprove it empirically, too.  Given that empirics don't conclusively demonstrate either proposition, it seems better to go with the one that it is theoretically consistent with individual rationality rather than the one that assumes people act like robots, spending on the basis of their "marginal propensity to consume" without thinking about the future.  Empirics have invalidated this view of consumption, by the way (see Friedman, 1957, described here). 

Keynes, in The General Theory, deployed his literary gifts to laugh to scorn the "classical" view (to some extent a strawman) against which he posed as a revolutionary; his argument was basically that the obvious existence of unemployment in the world disproved the classical theory.  But there are, in general, unemployment stories consistent with classical macroeconomics (e.g., efficiency wages).  Anyway, to argue that unemployment in 1936 was a failure of laissez-faire capitalism was absurd to the point of dishonesty, given that massive tax hikes and an explosion of invasive government policies had in the past few years put capitalism under siege.  The weirdest thing about Keynes' General Theory is that he either ignores or dismisses the obvious explanation why business investment was weak in the 1930s, namely that high taxes, extensive and unpredictable interference in the private sector by government, and FDR's vilification and arbitrary prosecutions of leading businessmen had destroyed the incentive to invest.  Here Keynes, a socialite and a snob, sacrificed logic to political correctness.  A 1930s version of Obamamania, far more virulent than our present affliction, made it intolerably gauche to oppose the progressive hero of the times, FDR.  For young economists, Keynes provided an excuse to shirk the irksome duty of exposing the harmful consequences of FDR's policies, and they embraced his ideas with enthusiasm.

Wolf thinks it's a good sign that Treasury rates are rising.  He argues that the fall in Treasury rates was a result of a panic, and the recovery of Treasuries is a sign of recovery.  Of course, there could be a good reason for Treasury rates to rise, namely, if business investment was swinging into action again and the rise in Treasury rates reflected competition for loanable funds from hungry private borrowers.  But how can we plausibly guess that this is the reason for the rise in Treasury rates when there is a far more obvious reason, namely that the government is borrowing so much that it is driving up rates, combined with growing worries about the solvency of the US government?  The causal link is from government spending to higher interest rates to crowding out of private business investment and exacerbation of the problems of homeowners with adjustable rate mortgages.  Which is why Keynesian fiscal policy is a bad idea right now.  (A limited amount of counter-cyclical fiscal policy, e.g., "automatic stabilizers," might be justified in normal recessions, and could be justified on grounds that don't appeal to the Keynesian intellectual apparatus.)

"History lessons for economists in thrall to Keynes" (Niall Ferguson, FT)

On Wednesday last week, yields on 10-year US Treasuries – generally seen as the benchmark for long-term interest rates – rose above 3.73 per cent. Once upon a time that would have been considered rather low. But the financial crisis has changed all that: at the end of last year, the yield on the 10-year fell to 2.06 per cent. In other words, long-term rates have risen by 167 basis points in the space of five months. In relative terms, that represents an 81 per cent jump.

Most commentators were unnerved by this development, coinciding as it did with warnings about the fiscal health of the US. For me, however, it was good news. For it settled a rather public argument between me and the Princeton economist Paul Krugman.

It is a brave or foolhardy man who picks a fight with Mr Krugman, the most recent recipient of the Nobel Prize for Economics. Yet a cat may look at a king, and sometimes a historian can challenge an economist.

A month ago Mr Krugman and I sat on a panel convened in New York to discuss the financial crisis. I made the point that “the running of massive fiscal deficits in excess of 12 per cent of gross domestic product this year, and the issuance therefore of vast quantities of freshly-minted bonds” was likely to push long-term interest rates up, at a time when the Federal Reserve aims at keeping them down. I predicted a “painful tug-of-war between our monetary policy and our fiscal policy, as the markets realise just what a vast quantity of bonds are going to have to be absorbed by the financial system this year”.

De haut en bas came the patronising response: I belonged to a “Dark Age” of economics. It was “really sad” that my knowledge of the dismal science had not even got up to 1937 (the year after Keynes’s General Theory was published), much less its zenith in 2005 (the year Mr Krugman’s macro-economics textbook appeared). Did I not grasp that the key to the crisis was “a vast excess of desired savings over willing investment”? “We have a global savings glut,” explained Mr Krugman, “which is why there is, in fact, no upward pressure on interest rates.”

Keynes had some genuine, albeit marginal, contributions to make, but where he posed as an antagonist of traditional economics, tradition was generally vaguely right and Keynes was generally guilty of logical or empirical errors.  That economists made Keynes the most influential economist of the 20th century was just a mistake on their part.  Yes, deficits do raise interest rates.

"Summers, Geithner Are Silent as IMF Loses Grip" (Bloomberg):

The “Asian Monetary Fund,” so passionately derided by Summers and Geithner [in 1997/8], is back. There is little a crisis-plagued U.S. can do to stop Asia’s $120 billion foreign- exchange reserve pool. Its creation is the clearest sign yet that Asia's getting serious about combating the global crisis. Done right, it will bode well for the region’s outlook...

[Summers' and Geithner's] concern [in 1997/8] was that [an Asian Monetary Fund] would eclipse the IMF in the region and, by extension, the U.S.’s say in how Asia retooled economies. They feared doling out billions of dollars in aid with few policy-change strings attached would prove dangerous. And they got their way. That changed this week.

From 1997 to 1998, the IMF arranged about $100 billion of loans to Indonesia, South Korea and Thailand as currencies collapsed. In return, governments had to cut spending, raise interest rates and sell state-owned companies. The IMF’s policies needlessly deepened the turmoil. 

Aside: Asia recovered quickly under IMF "austerity" policies.  I doubt we'll recover so fast.  If only we could be under the stern tutelage of the 1997-8 IMF instead of under the Obama administration.

During this global crisis, nations sought help from neighbors instead of borrowing from the IMF.  Indonesia, for example, raised $5.5 billion of standby loans from theAsian Development Bank, World Bank, Australia and Japan, and increased the size of its currency-swap arrangements with China and Japan to bolster access to foreign exchange.

Asia’s new fund takes things to another level. On May 3, the Association of Southeast Asian Nations, together with Japan, China and South Korea, agreed on terms for the so-called Chiang Mai Initiative and to use the funds in times of trouble. The pool, to be ready by year-end, largely neuters the IMF in Asia.

Granted, $120 billion seems like a paltry sum in today’s world of cascading markets and recession. American International Group Inc. alone received $183 billion worth of public bailouts. The size of the stockpile isn’t the issue. It can always get bigger.  China and Japan boast almost $3 trillion of currency reserves. It’s the symbolism that really matters here.

Asia as been lacking in the cooperation department as the U.S. redit crisis infected global markets. Meeting in Bali, Indonesia, this week, governments put their money where their vulnerabilities are, and it’s about time.

Nice to hear of initiatives originating elsewhere.  Somebody needs to take the reins of world leadership now that we're falling apart.

College and Redistribution

I was thinking today about the redistributive effects of government subsidies to education, and it occurred to me that there are two quite different ways of looking at it.

First, think about a young cohort.  Some go to college, some don't.  Those who go to college will end up richer.  So to subsidize college is, in the long run, to redistribute to the rich.  Presumably that's undesirable.

But then, think about an older cohort, the parents of a college age cohort.  Some of these people have kids, some don't.  Those who have raised kids will have incurred a lot of costs creating the new generation whose taxes will sustain the public finances when this cohort is old.  A college subsidy charges all taxpayers to relieve the costs for parents, thus redistributing from the childless to parents.  That seems desirable for two reasons.  First, the childless are likely, ceteris paribus, to be better off financially since they've had fewer expenses.  Second, it encourages people to have kids rather than being childless, which is good for the future tax base.

So do subsidies to college education redistribution in a perverse way, or in a desirable way?

This blog post from Robert Reich has some facts correct, but this dismissal is implausible to the point of dishonesty:

"The huge debts we're wracking up will cause your taxes to rise!" Wrong again. When it comes to the national debt, as I've said before, the relevant statistic is the ratio of debt to the gross domestic product. The only sure way to bring that debt down and make it manageable in future years is to get the economy growing again -- which requires that, in the short term, the government spend a lot of money (because consumers and businesses won't). In the long term, the biggest source of concern is rising health-care costs. And that's something Obama and Congress are aiming to tackle.

I have yet to hear of any strong non-military examples of the government actually stimulating the economy by spending lots of money. Two partly-military examples of effective stimulus, the 1960s and the Bush boom of 2003-2006, ended in disasters, namely the 1970s stagflation and today's crisis. Japan's experience started in disaster and stayed, while East Asia, which tried the opposite approach, austerity, under IMF pressure, rebounded quickly and has been flourishing ever since. This pattern is not too surprising in a way. There are trade-offs. If you manage the macroeconomy for the short run, you hurt yourself in the long run. Ultimately, this is just Adam Smith. The invisible hand of the market arranges things better than the visible hand of government. If government starts doing more, we get poorer. The current spending explosion probably lowers the expected time-path of GDP.

But even if it didn't, it's still a lot of added debt, and you have to pay that back somehow. Reich should say, "Yes, your taxes are going up, but it's worth it," not pretend that government spending is a free lunch.

Why Military Spending Provides the Best Stimulus

In the 1930s, FDR engaged in a lot of deficit spending. It didn't really work. One might argue that it partly worked, since there was some recovery from 1933-37. I would argue that that was because of banking reforms and expansionary monetary policy in those years, and that the deficit spending probably did less than nothing to help the recovery. In any case, the economy at its peak in 1937 was still below its 1929 levels, the only time in American history that the peak of one business cycle has fallen short of the peak of the previous one. Fiscal policy worked feebly at best.

Then came WWII, when the government borrowed and spent like crazy and the economy-- at least, according to GDP statistics-- came roaring back. Advocates of fiscal stimulus these days never cite FDR's deficit spending in the 1930s, they cite WWII in favor of their policy recommendation. And it does seem that there have been several episodes in which military spending has served as an economic stimulus. It did so in Hitler's Germany starting in 1933, when Hitler spent heavily on rearmament. It seems to have done so in the United States in the 1960s, as the Vietnam War helped us through a long-lasting boom. And an increase in deficit defense spending may have fueled the comparative (at least, to today!) prosperity of 2003-2006. I am aware of no examples of successful Keynesian fiscal stimulus that were not based on military spending. There is at least one apparent massive failure: Japan in the 1990s, where fiscal stimulus bought only a decade of stagnation and put the country up to its neck in debt.

If this empirical pattern is accurate, what is the explanation? Why does military spending work better than other kinds of spending? I can think of three reasons:

1. Military spending systematically distorts measured GDP. In ordinary times, GDP may be considered to be a decent indicator of human welfare, since it is built on market prices that reflect the "revealed preferences" of individuals. If the economy produces a lot of new stuff that is basically worthless, that raises Soviet GDP since widespread fixed prices and queues prevent people from revealing their real preferences, but it wouldn't raise US GDP because the stuff would sell for next-to-nothing if at all, and would be priced into GDP accordingly. But military spending is typically not mediated through the market, or at most through highly restricted and politicized markets like defense contracting. Money-value becomes disconnected from utility theory and revealed preference. GDP was higher during WWII in large part because we were producing lots of planes and bombs and warships etc. which we priced at certain levels, but these things weren't making anyone any happier (at least, not in the way potatoes or pizzas or pianos or palaces do). By this account, the effectiveness of military fiscal stimulus is a statistical illusion.

2. During wartime, the rights of workers tend to be restricted, enabling the government and employers to work them harder. Certainly in Hitler's Germany, the rights of labor were curtailed. I think it's also the case that union activism died down in the US in the 1940s relative to the 1930s. And the draft is a curtailment of the rights of workers which has no parallel in peacetime. I'm not saying any of this is good, but it might tend to raise GDP.

3. This is the most important. Defense spending is temporary. A general problem with Keynesian economics is that it assumes policy is made by farsighted and disinterested technocrats, whereas in reality it is made by politicians beholden to all manner of special interests and none too bright themselves and looking for short-term electoral gain and partisan reputation. Once a program is introduced, it creates a constituency for itself and is then difficult to reverse, no matter how obsolete its original rationale becomes. An increase in long-term government spending, if we make the sensible if suddenly unfashionable assumption that government spending is less efficient than private spending, will tend to reduce GDP. An increase in short-term government spending when resources are unemployed might, in theory, increase GDP by employing some of these resources. But if a given increase in government spending is expected to be long-term rather than short-term it will be more likely to crowd out private spending than to increase overall spending (long Keynesian story behind that, and it could be disputed on various grounds, but I say it for the sake of argument).

What is unique about war is that, unlike all other forms of government spending, it is inherently and credibly temporary. The Iraq war is possibly an exception to this since it seems to be the expression of a new foreign-policy doctrine which implies more war than previously expected. Still, war does not create a constituency for its continuation in the same sense that other government programs do, for the simple reason that it is unpleasant, and most unpleasant for the "beneficiaries" in a narrow fiscal sense, namely, the soldiers who get paid (not very much) to do the fighting and dying in it. Of course, defense contractors and congressional districts hosting military bases may have an interest in keeping war going. But it's reasonable for private sector investors to expect, "We'll fight for a while, we'll win the war, and then we'll come home and spending will fall."

With other government spending, though they may have some tendency to be cyclical, the best predictor of future levels of spending may be present levels of spending plus the trend in the level of spending. An investor thinks: "Uh-oh, we're spending a lot more now, probably we'll spend that much more forever. So we're all poorer. I guess I'd better spend less, in order to save more and be ready for this poorer future."