Larry Summers’ "Why America Must Have a Fiscal Stimulus"

In the Financial Times, former Treasury Secretary Larry Summers calls for the implementation of a fiscal stimulus package in the US in the first half of 2008. He argues that the question is not whether to prop up the economy, but how, that is, whether via monetary or budgetary actions. His view is that a “diversified” approach is better than relying on monetary policy alone, since government spending or tax relief programs can give aid to families that are hurt by a slowdown, while monetary policy’s direct impact is on financial institutions. Summers also believes that government measures can shore up faltering consumer demand.

Now I hate to sound like an economic Luddite, but I question this preoccupation with avoiding a recession. It appears to be based on wishful thinking that has been around since the 1960s, that the economy can be tuned so finely as to make slowdowns a thing of the past.

Yes, recessions result in job losses, which tend to hit lower income workers worse (although, ironically, this time around, workforce cuts are likely to be greatest in financial services, which pays above average wages), so it would be preferable to avoid them. But are we kidding ourselves as to the cost/benefit tradeoff?

Consider this selection from a recent post on the Fed’s Jackson Hole conference:

James Hamilton (enough of a Serious Economist to get to present a paper as Jackson Hole) comments approvingly on an observation by UCLA’s Ed Leamer (note he was lukewarm about other aspects of Leamer’s presentation):

I found another of Leamer’s main themes to be an intriguing suggestion. He claims we should think of monetary policy as doing very little about the long-run growth rate (which he thinks will be within 3% of a 3% annual growth line regardless of policy), and that stimulating the housing market therefore just changes the timing. Specifically, Leamer believes we bought ourselves a boom in 2004-2006 at the expense of a recession in 2007-2008.

That view suggests that monetary stimulus is no free lunch.

Now what if Leamer is right, that cheap credit pushed the US above trend-line growth and a period of below-average growth is inevitable? That means that the best stimulus measures can do is reduce the severity of the slowdown but at the cost of increasing its length. At worst, if they succeed in pushing growth to or above trend line, they will make the inevitable contraction worse.

So the real problem may be that we want to have our cake and eat it too. There is some evidence that a service based economy will show lower productivity gains than a manufacturing-driven economy (remember, economic growth is due to population gains and productivity improvements). But high growth periods help assure re-election, among other things. So the public at large approves of the good times they enjoy in unsustainable high growth periods, and then wants to avoid the inevitable consequences of a retrenchment.

And if you subscribe to the Schumpeterian line of though, recessions are a useful, “creative destruction” phase. They set the foundation for future growth, thining the herd and punishing excesses. And we’ve seen plenty of excesses recently.

Summers also argues that use of fiscal rather than monetary measures will reduce pressure on the dollar, since rates will not be cut as far. Short-term, yes, but remember, the weakness of the dollar is ultimately due to our negative domestic savings. And Summers tells us that one reason to implement a stimulus package is to keep aggregate demand up. Presumably, consumers will go from near zero savings to a modest level of savings, but that will be matched (or even more than offset, given job losses) by an increase in the government deficit. Our negative savings will increase, which will lead to an increase in our current account deficit, which will depress the dollar. Again, no free lunch.

In fairness to Summers, things may get so bad that some sort of stimulus package becomes necessary. And it is generally believed that stimulus is more effective when applied early rather than late. So if the US is going to go down the path of increasing the deficit in an effort to salvage the economy, better to do it sooner rather than later.

From the Financial Times:

The odds of a 2008 US recession have surely increased after a very poor employment report, growing evidence of weak holiday spending, further increases in oil prices, more dismal housing data and further writedowns in the financial sector. Six weeks ago my judgment in this newspaper that recession was likely seemed extreme; it is now conventional opinion and many fear that there will be a serious recession. Markets now predict the Federal Reserve will provide further stimulus to the economy by cutting rates by an additional 125 basis points on top of the 100 basis points they have already been cut so that rates fall to the 3 per cent range.

Given the market’s prediction of Fed policy actions, the debate now is not about whether or not to provide macro economic stimulus. That question appears to be settled. The question is whether it is better for all the stimulus to come from discretionary monetary policy or for some of the stimulus to come from discretionary fiscal policy. A diversified policy approach seems clearly preferable in that (i) in a world where judging the impact of policy measures is difficult, the outcome is less uncertain with a diversified mix of stimulus measures; (ii) the proximate impact of fiscal policies is felt by the families bearing the brunt of recession, in contrast to monetary policies whose immediate impact is on financial institutions; (iii) use of fiscal policy reduces the amount by which interest rates have to be reduced, thereby reducing downward pressure on the dollar, which in turn contributes to upward pressure on US inflation and international instability; (iv) partial reliance on fiscal policy mitigates the various risks of bubble creation associated with excessively low interest rates.

Beyond policy mix considerations there is the desirability of maintaining stable demand by insuring against excessive declines in consumer spending that lead to reduced employment and further declines in incomes and spending. The economy has been more stable in recent years than historically – one reason is that consumer credit markets have allowed households that suffered income declines as the economy turned down to maintain spending by borrowing on credit cards or home equity. These mechanisms, like monetary policy, are less reliable with burdened borrowers and troubled financial institutions. Japan’s experience in the early 1990s – when it failed to act decisively to respond to a downturn associated with collapsing financial bubbles and then experienced a disastrous vicious cycle of economic downturns and credit problems – should be highly cautionary regarding the importance of supporting consumption in the wake of financial problems.

Fiscal stimulus is appropriate as insurance because it is the fastest and most reliable way of encouraging short run economic growth at a time when a serious recession downturn would pressure American families, exacerbate financial strains, raise protectionist pressures and hurt the global economy.

Poorly provided fiscal stimulus can have worse side effects than the disease that is to be cured. This suggests close attention to three issues:

First, to be effective, fiscal stimulus must be timely. To be worth undertaking, it must be legislated by the middle of the year and be based on changes in taxes and benefits that can be implemented almost immediately.

Second, fiscal stimulus only works if it is spent so it must be targeted . Targeting should favour those with low incomes and those whose incomes have recently fallen for whom spending is most urgent.

Third, fiscal stimulus, to be maximally effective, must be clearly and credibly temporary – with no significant adverse impact on the deficit for more than a year or so after implementation. Otherwise it risks being counterproductive by raising the spectre of enlarged future deficits pushing up longer-term interest rates and undermining confidence and longer-term growth prospects.

Taken together these criteria suggest the desirability of a programme of equal payments to all those paying either income or payroll taxes combined with increases in unemployment insurance benefits for the long-term unemployed and food stamp benefits. Such a programme could be implemented quickly and would largely benefit those most likely to be cut off from credit markets and with the most urgent need to spend. It could easily be made temporary. Ideally, further stimulus would be provided by measures to reduce future deficits and increase long-run confidence.

How large should a programme of fiscal stimulus be? It depends on what else is done to help the economy – a subject to which I will return soon. But a $50bn-$75bn package implemented over two to three quarters would provide about 1 per cent of gross domestic product in stimulus over the period of its implementation. With some multiplier effects the total impact would be in the range of 1 per cent of GDP over a year. This seems large enough to take some burden off monetary policy and yet unlikely, if properly implemented, to risk substantial damage given flexible monetary policy if the economy proves stronger than expected. After many months of behind-the-curve policy, moving to implement such measures seems more prudent than waiting till the necessity of even greater ones has been unambiguously established by further pain.

Print Friendly, PDF & Email


  1. burrite

    I don’t mean to appear cynical, but what world are Prof Summers & the FT living in where they expect the Bush Administration to swiftly deliver targeted fiscal stimulus in the form of $50-$75 bn of extended unemployment and food stamp benefits to the people hit hardest by the economic slowdown? In 2001, when the evidence of a recession was clearer than it is today, their idea of targeted fiscal stimulus was to phase out the inheritance tax. That, of course, is why most observers are looking toward monetary policy as the primary tool in the current case.

    I’m also not sure I share much sympathy with the view that “a bit of recession may be a good thing now and then.” The last recession in ’01 and the one that seems to have started last month had similar causes, in my view: speculative bubbles had caused an excessive amount of capital to be directed toward the bubbly sectors (telecom bandwidth the last time and housing/mortgages this time). The only credible reason to allow a recession (assuming the Fed & the Treasury had the power to avoid one) would be to make sure that the market got the message that too much capital had been allocated to the offending sectors. But surely the market both last time & this time seems to have gotten the message just fine. You didn’t see any venture-backed long-haul optical startups in 2001-2, and you are unlikely to see many major residential real estate projects initiated in ’08-’09.

    Just my two cents.

  2. Yves Smith


    Agreed completely on your first point.

    On your second, I may be seeing things too much from an NYC perspective, but despite the noise, the 2001 recession was tame (if you lived outside Silicon Valley or were not otherwise getting a lot of dot-com related business. Management consultants, for instance, were hard hit). The 1980-1981 and 1991-1992 recessions were nasty, scary affairs. The 1991-1992 recession was so bad that Goldman partners were retiring early (that turned them into limited partners and froze the value of their holdings). It was a huge vote of no confidence. And in historical Wall Street downturns, the peak to trough fall in industry employment has been 20%. We didn’t see anything close to that in 2001.

    So the cheap credit extended to keep the recession from going to what is arguably its needed level instead leads as you pointed out to asset inflation. And the sort of people who invested in dot com stocks were sufficiently different than the types that invested in real estate debt that very few people at the time connected the dots.

  3. Gavin


    I’m sick and tired of hearing about these proposed stick saves.

    The biggest financial stimulus package that could be put together is a comprehensive set of regulations that force one or more I-banks to declare bankruptcy (Citigroup, I’m looking at you) and the rest to write everything off to get on with it.

    Because this is impossible, we won’t see a recovery until at least mid-09.

    Everything else is throwing good money after bad. You don’t fix the problems stemming from past easy money with more easy money.

  4. Independent Accountant

    I’m with you. For my money, Yves Smith is right and Larry Summers is “talking trash”. We need a number of major financial institutions to fail. As it is, the only one I think the Fed will let fail is Countrywide. As Milton Friedman told us many times, there is no such thing as fiscal stimulus. The stimulus is in how the increased deficit is financed.
    Yes, the recession will he harder in NYC as $5 million a year investment bankers seek real work. Also: California, Arizona, Florida and Nevada.

  5. tedb

    I agree with Burrite that the Bush administration had its chance to implement an appropriate fiscal response in 2001 and they not surprisingly blew it. Summers is correct to argue for a fiscal stimulus that is timely, targeted (to the poor and middle class), and temporary. The Bush administration only wants permanent relief to the rich.

  6. alex

    I agree with those who believe that recessions are necessary to cure excessive mal-investment in a specific sector (simply put, an asset bubble). However, I also assert the point that some (most notably Gavekal) have made in distinguishing between “good” bubbles, in which NEW productive assets are created (eg.the worldwide network of fiber-optic cable laid in the tech/TMT bubble) and “bad” bubbles.

    The recent credit bubble can be seen as particulary “bad” because loose credit enabled a massive allocation of capital to Non-productive assets, ie, houses. The debt to purchase these assets is by definition, non self liquidating. Combine that with the size of the asset class and its importance as the primary investment of the average american consumer, and we can see why this cycle of mis-allocation of capital may be particularly damaging to future growth.

    We need to see some Schumpeterian “destruction” — of banks, of builders, of hedge funds, etc. — if only to promote the idea that greater regulation and/or transparency needs to be implemented in both Mortgage Lending and Structured Finance so that we aren’t able to convince ourselves of the same beliefs that fueled this mania for “investing” in non-productive assets.

    We should only consider fiscal policy action if we feel that the threat of systemic risk has become so large that it dwarfs any consideration of prolonging the recession or incentivizing moral hazard.

  7. Anonymous

    As income and wealth numbers show, there are at least two economies in America (and, in theory, possibly a third; namely, those on the brink between just eking it out as participants in the ‘doing fine’ economy versus falling into the abyss of the ‘can’t make ends meet economy’).

    This matters to any discussion of Summers’ proposal. We ought to take time to reflect on the pro’s and con’s of fiscal and monetary policy with regard to each of the two economies. We also ought to opine on the pro’s and con’s of recession on each of the two economies. And, finally, we ought to reflect, as per alex, on the nature of asset bubbles as they affected/affect the two economies.

    Let’s start with this: the ‘can’t make ends meet’ economy in which at least 150 million Americans already inhabit has been in serious recession for a long time. We’re only getting discussion about the ‘recession’ and, for example, the illusory government statistics (GDP, employment, inflation, poverty level) that have been used to mark the first economy to model because, well, you know why: folks are looking to some bad bonuses.

    The housing/credit bubble sucked cash flow out of the distressed economy as well as the third group on the edge. Indeed, it drove many of the third group over the edge. The gains from sucking blood from a stone (what yves calls ‘zero to negative savings”; but we must also call: borrow and consume or perhaps, ‘livin large’) — these gains went to pump up the fees, salaries, benefits, and balance sheets of the first economy. As per the split in GDP gains between capital and labor, they certainly didn’t go to jobs or wages. Indeed, that was key to the strategy: instead of wages to support economic growth, we have seen borrowing — or put differently: Beggar thy neighbor (Only, of course, most of the first economy folks live in gated off ways so that they don’t have any sense of second/third economy folks being ‘neighbors’. Compare, for example, the immigration stands of first economy folks versus rest of nation.)

    Further, because the first economy is knitted into the upside of the global economy — these gains also allowed the wizards of financial innovation and the very wealthy folks to whom they report to help sell out long term American economic security in exchange for fabulous personal gains.

    In comparison, the tech bubble printed counterfeit stock to make lots of people rich in a similar fashion, but as alex points out, also to go on a wild R&D spree that, at least as a byproduct, led to new technologies that, in theory, could benefit others later.

    This is just one comparison. Many more along all the various lines are needed to best understand what would be the ‘right’ thing to do. Not that Bush — or the economic elite who live in the first economy will ever do the right thing. But at least so that we can see what might make sense should we ever regain leadership with foresight and courage who care about more than themselves.

Comments are closed.