The year 2013 marks the 100^th anniversaries of two separate major institutional innovations in American economic policy:  the Constitutional Amendment enacting the federal income tax, ratified on February 3, 1913, and the law establishing the Federal Reserve, passed in December 1913.  
           It took some time before the two new institutions became associated with the explicit concepts of fiscal policy and monetary policy, respectively.   It wasn’t until after the experience of the 1930s that they came to be viewed as potential instruments for managing the macro-economy.  John Maynard Keynes, of course, pointed out the advantages of expansionary fiscal policy in circumstances like the Great Depression.   Milton Friedman blamed the Depression on the Fed for allowing the money supply to fall.    [Tools of fiscal policy used by governments, in addition to tax rates and tax deductions, are spending and transfers.  Tools of monetary policy used by central banks include interest rates, quantities of money and credit, and instruments such as reserve requirements and foreign exchange intervention used in various (non-US) countries.]

           In subsequent debate, Keynes was associated with support for activist or discretionary policy.  The aim was counter-cyclical response to economic fluctuations: expansion in recessions, discipline in booms.  (It is a myth that he favored big government generally.  He said “the boom is the time for austerity.”)    Friedman opposed activist or discretionary policy, believing that government institutions, whether monetary or fiscal, lacked the ability to get the timing right.   But both great economists were opposed to pro-cyclical policy moves, such as the misguided US tightening of 1937 at a time when the economy had not yet fully recovered. 
          After World War II, the lessons of the 1930s were incorporated into all the macroeconomic textbooks and, to some extent, into the beliefs and actions of policy-makers.  But many of these lessons have been forgotten in recent decades, crowded out of public consciousness by experiences such as the high-inflation 1970s.  As a result, many politicians in advanced countries are repeating the mistakes of 1937 today.  This despite conditions that are qualitatively similar to those that determined Keynes’ policy recommendations in the 1930s: high unemployment, low inflation, and rock-bottom interest rates.

         The austerity-versus-stimulus debate has been thoroughly hashed out.   On the one hand, proponents of austerity correctly point out that the long-term consequences of permanently expansionary macroeconomic policy [both fiscal and monetary] are unsustainable deficits, debts, and inflation.    On the other hand, proponents of stimulus correctly point out that in the aftermath of a recession, when unemployment is high and inflation low, the immediate consequences of contractionary macroeconomic policy are continued unemployment, slow growth, and debt/GDP ratios that go up rather than down.  Procyclicalists, both in the US and Europe, represent the worst of both worlds:  they push in the direction of expansion during booms such as 2003-07 and in the direction of contraction during recessions such as 2008-2012, thereby exacerbating both the upswings and downswings.  Countercyclicalists have it right:  working in the direction of fiscal and monetary discipline during booms and ease during recessions.

           Less thoroughly aired recently is the question whether — given recent conditions - monetary or fiscal expansion is the more effective instrument.   This question was addressed clearly in 1937 by Sir John Hicks in a once-famous article titled “Mr. Keynes and the Classics.”  The graphical model is known to many generations of undergraduate students in macroeconomics under the label “IS-LM.”   
           The answer to the question which form of policy is more effective:  under the circumstances that held in the 1930s and that hold again now - which are conditions not just of high unemployment and low inflation, but also near-zero interest rates — stimulus in the specific form of fiscal expansion is much more likely to be effective in the short-term than stimulus in the form of monetary expansion.   Monetary expansion is rendered relatively less effective because interest rates can’t be pushed below zero.  (Flat LM curve.)  This situation, labeled by Keynes a liquidity trap, is today called the Zero Lower Bound.  In addition, firms are less likely to react to easy money by investing in new plant and equipment if they can’t sell the goods they are producing in the factories they already have.  (Steep IS curve.)  The hoary — but still evocative — metaphor is “pushing on a string.”  Meanwhile, fiscal expansion is rendered relatively more effective than in normal times, in that it doesn’t push up those rock-bottom interest rates and thereby crowd out private-sector demand.
           Despite the inability of central banks to push short-term nominal interest rates much lower, one should not give up completely on monetary policy, especially because fiscal policy is so thoroughly hamstrung by politics in most countries.  It is worth trying all sorts of things:  quantitative easing, forward guidance, nominal targets.   Even if the short-term interest rate channel is inoperative, such steps may work through other channels:  long-term interest rates, credit channel, risk premia, expected inflation, asset prices, commodity prices or exchange rates.  But the effects of each are highly uncertain. 

          That monetary policy is less effective than fiscal policy under conditions of high unemployment and zero interest rates should not be a novel position.  But many economists have forgotten much of what they knew and politicians may not have even heard the proposition. 
          Introductory economics textbooks have long talked about the Keynesian multiplier effect:  the recipients of federal spending (or of consumer spending stimulated by tax cuts or transfers) respond to the increase in their incomes by spending more as well, as do the recipients of that spending, and so on.  Again, the multiplier is much more relevant under current conditions than in the normal situation where the expansion goes partly into inflation and interest rates and thus crowds out private spending.  By the time of the 2008-09 global recession even those who believed that fiscal stimulus works had marked down their estimates of the fiscal multiplier — intimidated, perhaps, by theories of policy ineffectiveness.   (These are some of the same theories that predicted that a tripling of the monetary base over five years, or a near-doubling of M1, should double or triple the price level !)
          The subsequent continuing severity of recessions in the United Kingdom and other countries pursuing contractionary fiscal policies, apparently to the surprise of the politicians enacting them, suggested that fiscal multipliers are not just positive, but greater than one, as the old wisdom had it.   The IMF Research Department has now reacted to this recent evidence and bravely confessed that official forecasts, including even its own, had been operating with under-estimates of multiplier magnitudes.
          A new wave of econometric research estimates fiscal multipliers using methods that allow them to be higher in some circumstances than others.   Baum, Poplawski-Riberio and Weber (2012) allow the estimate to change when crossing a threshold measure of the output gap.  Batini, Callegari and Melina (2012) allow regime-switching, across recessions versus booms.  Others that similarly distinguish between multipliers in periods of excess capacity versus normal times include Auerbach and Gorodnichenko (2012a, 2012b), Bachman and Sims (2012), Baum and Koester (2011), and Fazzari, Morley and Panovska (2012).  Most of this research finds high multipliers under conditions of excess capacity and low interest rates.  Gordon and Krenn (2011) and Shoag (2012) have the same implication.    Related studies confirm other conditions that matter for the size of the fiscal multiplier in precisely the way the traditional textbooks say, for example that they are lower in small open economies because of crowding out of net exports.  (Perhaps due to fear of sounding old-fashioned, few of these studies have the courage to mention that these are the findings that one would have expected from the elementary textbooks of 50 years ago.) 

          Needless to say, the effects of fiscal policy are subject to substantial uncertainty.   One never knows, for example, when rising debt levels might suddenly alarm global investors who then start demanding abruptly higher interest rates, as happened to countries on the European periphery in 2010.    (For this reason, the United States would be well-advised to lock in a long-term path toward debt sustainability, even while undertaking a little short-term stimulus.)   In the case of stimulus in the form of tax cuts, one never knows how much of the boost to disposable income will be saved by households rather than spent. We are also uncertain as to the magnitude of negative effects of high tax rates, via incentives, on long-term growth.   And it is true that monetary policy is much better understood than it was in the past. 
            Nevertheless, if the question is whether it is monetary policy or fiscal policy that can more reliably deliver demand expansion under current conditions, the answer is the latter.  One might even dramatize the contrast by speaking of “monetary alchemy and fiscal science.”

            A much-admired 2010 paper by Eric Leeper had it the other way around: it characterized monetary policy as science and fiscal policy as alchemy.   It is true that the state of knowledge and practice at central banks, which actually set the instruments of monetary policy, is close to the best that modern society has to offer.    It is likewise true that the instruments of fiscal policy are set in a very political process that is poorly informed by the state of economic knowledge and motivated largely by politicians’ desire to be re-elected.  These political realities may be what the author of “Monetary Science, Fiscal Alchemy” had in mind.
             But the ancient alchemists were not in fact stupid or selfish people in general, notwithstanding their search for the “philosopher’s stone” that was to turn lead into gold (of which modern proponents of returning monetary policy to the pre-1914 gold standard are reminiscent).  Nor was the alchemists’ problem that the monarchs of their day refused to listen to them.  It was rather that the state of knowledge fell far short of what the modern science of chemistry can tell us.   
            The term alchemy could be applied to pre-Keynesians like US Treasury Secretary Andrew Mellon (whose Depression prescription was that President Herbert Hoover should “liquidate labor, liquidate stocks, liquidate farmers, liquidate real estate… it will purge the rottenness out of the system”).  It could also be applied to the “Treasury view” in the UK of 1929. (Churchill:  ”The orthodox Treasury view … is that when the Government borrow[s] in the money market it becomes a new competitor with industry and engrosses to itself resources which would otherwise have been employed by private enterprise, and in the process raises the rent of money to all who have need of it.” ).  But in light of all that was learned in the 1930s, it would be misleading to characterize the current state of fiscal policy knowledge as alchemy.

References

   Miguel Almunia, Agustín Bénétrix, Barry Eichengreen, Kevin O’Rourke, and Gisela Rua, 2010, “From Great Depression to Great Credit Crisis: Similarities, Differences and Lessons,” Economic Policy, 25 (62), pp. 219-65.
   Alan Auerbach and Yuriy Gorodnichenko, 2012a, “Measuring the Output Responses to Fiscal Policy,” American Economic Journal: Economic Policy, vol. 4(2), pp. 1-27, May.
   Alan Auerbach and Yuriy Gorodnichenko, 2012b, “Fiscal Multipliers in Recession and Expansion,” NBER Chapters, in Fiscal Policy after the Financial Crisis, edited by Alberto Alesina and Francesco Giavazzi (University of Chicago Press).
   Rüdiger Bachmann and Eric Sims, 2012, Confidence and the transmission of government spending shocks,” Journal of Monetary Economics vol. 59, no.3, pp.235-249.  NBER WP No. 17063, May.
   Nicoletta Batini, Giovanni Callegari and Giovanni Melina, 2012. “Successful Austerity in the United States, Europe and Japan,” IMF Working Papers 12/190, International Monetary Fund.
   Anja Baum and Gerritt Koester, 2011, “The Impact of Fiscal Policy on Economic Activity Over the Business Cycle - Evidence from a Threshold VAR Analysis” Deutsche Bundesbank, Research Centre in its series Discussion Paper Series 1: Economic Studies  no. 2011,03.
   Anja Baum, Marcos Poplawski-Riberio and Anke Weber, 2012, “Fiscal Multipliers and the State of the Economy,” IMF Working Paper 12/286, International Monetary Fund, December.
   Olivier Blanchard and Daniel Leigh, 2013, “Growth Forecast Errors and Fiscal Multipliers,” IMF Working Paper No. 13/1, January.  Forthcoming, American Economic Review, May.  
   Steven Fazzari, James Morley, and Irina Panovksa, 2012, “State-Dependent Effects of Fiscal Policy,”  UNSW Australian School of Business Research Paper No. 2012-27, April.      
   Milton Friedman and Anna Schwartz, 1963,  A Monetary History of the United States, 1867-1960 (Princeton University Press).
   Robert Gordon and Robert Krenn, 2011, “The End of the Great Depression 1939-41: Policy Contributions and Fiscal Multipliers,” NBER Working Paper No. 16380.
   John Hicks, 1937, Mr. Keynes and the Classics: A Suggested Reinterpretation,” Econometrica, pp. 147-59.
   Ethan Ilzetzki, Enrique Mendoza & Carlos Vegh, 2011. “How Big (Small?) are Fiscal Multipliers?,” IMF Working Papers 11/52 (International Monetary Fund.)  Forthcoming, Journal of Monetary Economics.
   Eric Leeper, 2010, “Monetary Science, Fiscal Alchemy,” NBER Working Paper No. 16510.
   Christina Romer and David Romer, 2013, “The Most Dangerous Idea in Federal Reserve History: Monetary Policy Doesn’t Matter,” UC Berkeley, January.
   Daniel Shoag, 2012, “The Impact of Government Spending Shocks: Evidence on the Multiplier from State Pension Plan Returns,” Harvard Kennedy School.
   Antonio Spilimbergo, Steven Symansky, and Martin Schindler, “Fiscal Multipliers,” Staff Position Note No. 2009/11, International Monetary Fund.

[This post appears at VoxEU.  And also at  Econbrowser. Comments may be posted there.]

With November’s election fast approaching, the Republican candidates seeking to challenge President Barack Obama claim that his policies have done nothing to support recovery from the recession that he inherited in January 2009. If anything, they claim, his fiscal stimulus made matters worse.  And, despite recent improvement, the level of unemployment indeed remains far too high.not blame George W. Bush for the recession that began two months after he took office in 2001. There hadn’t yet been time for bad policies to damage the economy.)

Obama’s Democratic defenders counter that his policies staved off a second Great Depression, and that the US economy has been steadily working its way out of a deep hole ever since.  Middle-ground observers, meanwhile, typically conclude that one cannot settle the debate, because one cannot know what would have happened otherwise.

There is a good case to be made that government policies - while not strong enough to return the economy rapidly to health — did indeed halt an accelerating economic decline.    By “government policies,” I mean not just the fiscal stimulus the new president steered through Congress when he took office, but also the Obama version of TARP, and Fed Chairman Ben Bernanke’s aggressive monetary stimulus.   All three policy initiatives remain extremely unpopular with Republicans, and ambiguous among swing voters.

But the middle-ground observers are of course right that one cannot prove what would have happened otherwise.   It is also true that it is rare for a government’s policies to have a major impact on the economy immediately.  These things usually take time.  One cannot infer the merit of a new president’s policies from the path of the economy during his first few months in office.  (For example, I did

But here is the remarkable thing: whether one listens to the Republicans, the Democrats, or the middle-ground observers, one gets the impression that the economic statistics show no discernible improvement around the time that Obama took office. In fact, the reality could hardly be more different.

This is especially true if one looks at revised economic statistics, which show the US economy to have been in far worse shape in January 2009 than was reported at the time. In January 2009, the annualized growth rate in the second half of 2008 was officially estimated to have been negative 2.2%; but current figures reveal it to have been a horrendous negative 6.3%. This is the main reason why the level of economic activity in 2009 and 2010 was so much lower than had been forecast, which in turn explains why unemployment was so much higher.

Figure 1 shows the quarterly economic growth rate. The maximum rate of contraction — a veritable freefall in the economy — came in the last quarter of 2008 (the quarterly GDP data come from the Bureau of Economic Analysis of the U.S. Commerce Department).   More specifically, it came in December, according to the monthly GDP estimates from the highly respected MacroAdvisers.   (See monthly income figures in the form of growth rates in Figure 2 or levels of GDP in Figure 3.)  This was the month before Obama was inaugurated.  The situation miraculously began to improve as soon as Obama’s term began! 

(click here for larger graphs)

The full force of the fiscal stimulus package began to go into effect in the second quarter of 2009.    The NBER officially designates the end of the recession as having come in June of that year.  GDP growth turned positive in the third quarter.

US economic growth slowed down again in late 2010 and early 2011, as one can see in Figure 1.  The timing coincides with the beginning of withdrawal of the Obama fiscal stimulus. Indeed, the government has been the one sector to experience contraction in income and employment over the most recent five quarters.  The private economy has been expanding.

Other economic indicators, such as interest-rate spreads and the rate of job loss, also turned around in early 2009. Labor-market recovery normally lags behind that of GDP - hence the “jobless recoveries” of recent decades. But the graph of monthly job losses and gains reveals that here, too, the end of the freefall came precisely when Obama was inaugurated.  The last two charts show the same “V” shaped pattern in the monthly job change figures that are released by the Bureau of Labor Statistics, as the GDP growth figures that are released by the BEA.  The rate of job growth over the last two years, inadequate as it is, actually exceeds the rate of job growth during the Bush Administration, even if one counts only the period before the big recession hit in December 2007.

Again, these graphs do not demonstrate that Obama’s policies yielded an immediate payoff. In addition to the lags in policies’ effects, many other factors influence the economy every month, making it difficult to disentangle the true causes underlying particular outcomes.

What is the right way to assess whether the fiscal stimulus enacted in January 2009 had a positive impact?   Start with common sense. When the government spends $800 billion on such things as highway construction, teachers and policemen who were about to be laid off, and so on, it has an effect. Workers who would otherwise not have a job now have one. Furthermore, they may spend some of their income on goods and services produced by other people, creating a multiplier effect.

Those who claim that this spending does not boost income and employment (or that it even hurts), apparently believe that as soon as a teacher is laid off, a new job is created somewhere else in the economy, or even that the same teacher finds a new job right away. Neither can be true, not with unemployment so high and the average spell of unemployment much longer than usual.

They also think that the government deficit drives up inflation and interest rates, thereby crowding out other spending by consumers and firms. But interest rates are rock bottom, even lower than they were in January 2009, while core inflation is running at its lowest levels since the early 1960’s. The conditions of the last four years - high unemployment, depressed output, low inflation, and low interest rates - are precisely those for which traditional “Keynesian” remedies were designed.

Economists’ more sophisticated forecasting models also show that the fiscal stimulus had an important positive effect, for much the same reasons as the common-sense approach.   The non-partisan US Congressional Budget Office reports that the 2009 spending increase and tax cuts gave a positive boost to the economy, and indeed had the extra multiplier effects of the traditional Keynesian models. Allowing for a wide range of uncertainty [to allow for different economists' views], the CBO estimates that the stimulus added 1.5 percent to 3.5 percent to the level of GDP by the fourth quarter, relative to where it otherwise would have been.  The boost to 2010 GDP, when the peak effect of the stimulus kicked in, was roughly twice as great.

To be sure, of the many theoretical models produced by eminent macroeconomists at prestigious universities, some say that fiscal stimulus has no positive effect on the economy, even under recent economic conditions.  (The theoretical innovations underlyng the models have even won Nobel Prizes for the innovators, and not without justice.)  But these models are not sufficiently realistic to meet the market test:  they are not used by private businessmen for whom getting good forecasts matters to their planning and in turn to the success of their businesses.

Of course, econometric models do not much interest the public at large. A turnaround needs to be visible to the naked eye to impress voters. Given this, one can only wonder why basic charts, such as the 2008-2009 “V” shape in growth, have not been used - and reused - to make the case.

(Click here for larger versions of all 5 graphs.)

[Appears also at Fair Observer,with a nice presentation of the charts.

A shorter version appeared as an op-ed at Project Syndicate, which has the copyright.]

The National Journal asks: “Is the Obama administration’s stimulus plan helping to create or “save” 650,000 jobs, as the president and his aides say? Is that an appropriate way to measure the stimulus’ impact?”

My response:

I am astounded by claims that fiscal stimulus under recession circumstances doesn’t create jobs. Or at least I am astounded when such claims come even from some reputable economists.  Do they think that a construction job on a road-building project doesn’t count as a real job if the funding comes from the government?   More likely, they think that the increase in demand doesn’t raise output in the aggregate, because the federal debt crowds out private production and so someone else somewhere loses his or her job?  But that would be hard to believe, at a time when the Fed is keeping interest rates at zero, long-term interest rates are also quite low, and capacity is lying idle.  Moreover, Republican lectures to Democrats about the evils of the national debt take real chutzpah, after Presidents Reagan, Bush I and Bush II increased the debt ten-fold during periods when no national emergency required it.

Yes, the effort to identify specific jobs saved is more a political exercise than an economics exercise; but the true number of jobs saved, relative to what would otherwise have happened , is greater than the 650,000 number. It is legitimate as a communications strategy for the White House to make the benefits concrete by pointing to the many individual teachers who would have been laid off by fiscally devastated state and local governments in the absence of federal government money. But there are several difficulties with using this job count as a way to evaluate the impact of the fiscal stimulus.  One problem is that the exercise doesn’t count the indirect effects of most of the spending and tax cuts, where it is hopeless to try to pinpoint whose job was saved.

The biggest problem, of course, is that one cannot estimate accurately, let alone prove to the skeptics, what would have happened in the absence of the stimulus package. Claims by Republican congressmen that one should judge Obamanomics by looking at whether employment is greater now than before February are nonsense. If there hadn’t been a severe recession underway (starting on the predecessor’s watch, if you want to get political about it), there would have been no need for the stimulus. None of us claims that fiscal stimulus creates a lot of jobs on net when the economy is already expanding strongly. The increased government spending that occurred during the terms of Presidents Reagan and Bush after the recessions of their respective first terms had already ended, for example, did not create a lot of jobs.  But without the recent stimulus, the recession would have been worse.

The appropriate way to estimate the stimulus impacts is by means of a standard macroeconomic model with fiscal multipliers in it. But if you believe philosophically that fiscal multipliers are zero, even in a severe recession, then neither a standard macroeconomic model nor anything else will convince you.

(To post comments, go to the Roubini Global Economics version of this post.)

Personal saving rose again in the second quarter. “Does this mean the stimulus tax cut has failed, as the 2008 tax cut stimulus did?”, asks The National Journal.

My answer:

Martin Feldstein and others predicted that the tax-cut component of the 2009 fiscal stimulus package would have substantially less expansionary bang-for-the-buck than the spending component of the package, because much of the tax cut would be saved, as had been the case with the 2008 tax cut.  (“Bang for the buck” in this case could be defined as demand stimulus divided by budget cost.)   We knew this from Milton Friedman’s permanent income hypothesis, or even from good old Keynesian multiplier theory.

But in February President Obama had to get those last three (Republican) votes to pass the stimulus bill in the Senate, and those three Senators insisted on raising the tax cut component of the stimulus package a bit and lowering the spending component. Their motivation presumably was to mollify their fellow Republicans, many of whom still claim that ONLY tax cuts provide stimulus, and that spending does not (and perhaps even has a negative effect) — which is even more extreme than the claim that a tax cut creates stimulus equal to spending. After the failures of the Bush tax cuts (and Reagan’s before him), I don’t know if any economists still cling to such “supply sider” notions — or indeed if these congressmen would be able to state their logic. Regardless, I think the Feldstein prediction has been borne out since then.   Talk about irony!   The Reagan tax cuts of 1981-83 and the Bush tax cuts of 2001-03 were both explicitly designed to boost saving — hence their focus on capital income and higher income brackets — and yet in both cases private saving fell in their aftermath.   The tax cuts of January 2008 and February 2009 were both explicitly designed to boost consumption; yet private saving rose in their aftermath !   

Fortunately, the majority of the Obama stimulus package took the form of increased spending, much of which has yet to come.

None of this is to deny that efficiency is an important consideration, and cost-benefit calculations should always enter into the choice of both what kind of tax cuts to adopt and what kind of spending increases to adopt. But if it is short-term demand stimulus we are after, and we are, then government spending gives more bang for the buck than tax cuts.

[Any readers wishing to post a comment are encouraged to go to the versions on Seeking Alpha or RGE Monitor.]