Guest Contribution: “Biden Avoids Mistake of Insufficient Fiscal Stimulus”

Today, we present a guest post written by Jeffrey Frankel, Harpel Professor at Harvard’s Kennedy  School of Government, and formerly a member of the White House Council of Economic Advisers. A shorter version appeared at Project Syndicate.


It has been a year since the US and the world went into recession.  Because of its origins in the sudden pandemic, it was possible to reliably discern the advent of the recession before it was reflected in any of the standard economic statistics, which is rare.  (I hope it shocks no one to learn that economists can’t normally predict recessions.)

By the end of the second quarter of 2020, US GDP had fallen 11 %. This record plunge took the US economy from a level that is estimated to have been 1.0% above potential output at the end of 2019, to a level 10 % below potential in mid-2020.  Potential output is the level of GDP that is produced when unemployment is at its so-called natural rate, the capital stock is operating at the capacity for which it was designed, buildings have their normal occupancy rates, etc.

  1. Overheating?

Recently, the situation has changed dramatically.  Now forecasters expect US growth so rapid in 2021 that GDP will reach its pre-pandemic high very soon, and by 2022 will probably be above potential output.  (Recovery is also expected in the world economy, though not as rapid.)  Some economists now warn of US overheating.

This situation is, again, unusual:  It is rare enough for the economy to be 5 per cent above potential, as it last was in 1966; it is virtually unprecedented for such a rapid reversal to be predictable.  Only the US build-up to fight World War II is a possible exception.

It is not just the hoped-for victory of vaccination over the virus that is driving these rosy economic forecasts.  (“Hoped for,” not because of any fault of the vaccines, but rather because they are racing against an unholy alliance of anti-vaxers and mutations of the virus.)

US demand for goods and services is increasing very rapidly for several reasons.  First is pent-up demand. American households saved a good share (estimated at around $ 1 ½ trillion) of the government transfers that the Congress legislated a year ago and they are seen as spending some of it as soon as they can.

Second is the big monetary easing that the Federal Reserve put into place a year ago, including lowering short-term interest rates to zero.  Chairman Jay Powell has repeatedly pledged to leave them there for several years.

The third factor super-charging demand is the recently renewed fiscal relief. President Joe Biden’s $1.9 trillion American Rescue Plan, was passed earlier this month by a Democratic Congress. It came on the heels of $0.9 trillion in outlays legislated in December — as a final major act of the Donald Trump Administration, though he at one point threatened to veto the bill if direct payments were not raised to $2,000 per person.

A fourth boost to demand is planned, in the form of infrastructure investment spending.  Biden says that he will partly pay for this by raising taxes on corporations and the wealthy.  But raising taxes is much harder, politically, than raising spending.

No wonder Fed officials on March 17 sharply raised their projections for 2021 growth, to 6 ½ %.  The OECD did the same this month.

  1. Multiplying multipliers

The basis of the overheating assessments is easy enough that anybody can play, in round numbers.  Let us call the US fiscal expansion $1.9 trillion in 2021.  This is 9% of the total economy.

Keynesian multiplier theory is back in fashion, sufficiently so that one dares to speak its name. The multiplier is thought to be as high as 1.5 under recent conditions, namely, when interest rates are close to zero and inflation is low. (The CPI showed a rise of only 1.7 % from February 2020 to February 2021.)

When the government outlays take the form of income transfers, rather than spending directly on goods and services, only the part of the increase in disposable income that households consume enters the demand stream. They usually save some, as they certainly did in 2020, when a lot of the transfers went to the relatively well-off, who tend to save more than the poor.

So, call the multiplier 1.0.  Multiply that by the 9% of GDP in fiscal stimulus, and one gets a 9% increase in GDP.  At the end of 2020, the economy was an estimated 3% below potential output.  So, a 9% boost would put GDP about 6% (=9%-3%) above potential.  Even if the multiplier is only 0.5, it still puts the economy above estimated potential.

  1. Underestimation of potential output vs. flat Phillips curve

Some economists, most visibly former Treasury Secretary Larry Summers, while supporting the basic idea of Biden’s relief program, have on the basis of such calculations warned that the US economy is likely to overheat next year and that this would likely show up in the form of problematic inflation.  Financial markets have also reacted:  the interest rate on 10-year Treasury bonds has now risen to 1.7 %, up from 0.9% in January.

The counterargument is to point out that we don’t really know the level of potential output.  Maybe it is higher than the estimates.  Some have questioned whether, in retrospect, the economy in 2018-19 was really above potential after all, even though the unemployment rate fell as low as 3 ½ %, which is below conventional estimates of the natural rate of unemployment. Their evidence is that inflation rose very little, reaching only 2.3 % in 2019 in terms of the headline CPI.

It seems to me that the best explanation for the small magnitude of the rise in inflation in the period before the pandemic is less likely to be an underestimation of the level of potential output, and more likely to be what economists call a flat Phillips curve. That is, variation in employment and output has only small effects on wage and price inflation.

The evidence for the relative flatness of this relationship?  Before there was the puzzle of why inflation did not rise more than it did in 2018-19, there was the puzzle of why it did not fall more than it did during 2010-2015, in the aftermath of the Great Recession, when unemployment was coming down only slowly from its 10 % peak.  The explanation that encompasses both periods is that the curve is relatively flat.

The implication: yes, the US economy is likely to be above potential output next year; but, no, inflation is not likely to rise excessively.  For that matter, some rise in inflation is actively desired by the Fed, as part of the recovery effort.  Fed officials project the unemployment rate to fall to 3 ½ % by the end of 2023. Vice-Chair Rich Clarida on March 25 said the corresponding increase in inflation is consistent with the adoption of a new policy framework, Average Inflation Targeting, that the Fed adopted in August 2020.

Even Olivier Blanchard, who estimates that the fiscal boost will push unemployment as low as 1 ½ %, also estimates that the resulting rise in inflation would be just 0.5 percentage points in the short run.  (He worries about the longer run if the expansion continues.)

Other downsides?

To be sure there are other possible downsides to so much stimulus, besides inflation.
(1) The national debt is now at the highest level, relative to the economy, since 1945: the debt/GDP ratio is to reach well over 100% by the end of 2021, according to CBO.  It appears sustainable, provided interest rates remain very low.  But interest rates will rise sooner or later.  Ultimately, though debt dynamics constrain spending everywhere, the US can get away with things that other countries can’t, due to the dollar’s exorbitant privilege.

(2) Let’s say the Fed does succeed in keeping interest rates low. Many observers worry that such money-financed expansion causes asset bubbles and wealth inequality.

(3) The trade deficit is bound to increase, as some of the spending falls on imports.  In truth, this wouldn’t be so bad for the economy: the trade deficit would be a safety valve alleviating overheating.  But, politically, it could exacerbate protectionism.

(4) If “the sky’s the limit” on spending, some of the money could be wasted.  Fortunately, Biden is familiar with the possible pitfalls.  He has targeted much of the spending to such priority needs as fighting the pandemic and cutting child poverty and has taken credible steps to bolster accountability.

The “mistake of 2009”

The Biden people are clearly trying to make sure that the country doesn’t “repeat the mistake of 2009,” when Barack Obama’s $800 billion stimulus – though big by historical standards – was too little and too short-lived to do the job fully.   True, the Great Recession ended almost as soon as the legislation was put into place.  But the subsequent recovery was too slow.

It is not clear that the limited size of the 2009 American Recovery and Reinvestment Act was really a mistake of the Obama administration (unless, perhaps, the mistake was Obama’s desire for a modicum of bipartisanship). Its officials would argue that they got the biggest stimulus through Congress that was politically possible, given Republican opposition.

It may be off-limits to say it was the public who made a mistake.  But voters reflexively blamed the weakness of the 2009-10 recovery on the party that held the White House.  In the mid-term elections of November 2010, they put the Republicans back in charge of the House of Representatives, where they were in a position to block further measures to boost the economy.

The country did indeed make a mistake in 2009-10 in curtailing the size and duration of the fiscal expansion. Regardless of where precisely one locates that mistake, it is sensible of Biden to make sure it doesn’t happen again in 2021-22.


 This post written by Jeffrey Frankel.