Fiscal Policy in Depressions

Lawrence Summers and Brad DeLong have this paper which argues that in severely depressed economies, which are also constrained by the zero-interest rate bound, discretionary fiscal policy can be a powerful instrument to revive growth. They write,

In normal times central banks offset the effects of fiscal policy. This keeps the policy-relevant multiplier near zero. It leaves no space for expansionary fiscal policy as a stabilization policy tool. But when interest rates are constrained by the zero nominal lower bound, discretionary fiscal policy can be highly efficacious as a stabilization policy tool. Indeed, under what we defend as plausible assumptions of temporary expansionary fiscal policies may well reduce long-run debt-financing burdens. These conclusions derive from even modest assumptions about impact multiplier, hysteresis effects, the negative impact of expansionary fiscal policy on real interest rates, and from recognition of the impact of interest rates below growth rates on the evolution of debt-GDP ratios. While our analysis underscores the importance of governments pursuing sustainable long run fiscal policies, it suggests the need for considerable caution re-garding the pace of fiscal consolidation in depressed economies where interest rates are constrained by a zero lower bound.

Following the apparent triumph of monetarism in the seventies, Keynesianism had been upstaged as the dominant macroeconomic stabilization ideology for nearly three decades till the Great Recession took hold. It was believed that front-loaded fiscal consolidation for deficit-reduction coupled with accommodatory monetary policy would help achieve price stability, positively shape expectations and restore market confidence, encourage investment and consumption, and thereby boost aggregate demand. It would help successfully combat short-term business cycle problems and address medium-term growth dimensions.

It was also believed that the multiplier of discretionary fiscal policy was small. When the economy is close to its productive level, fiscal policy induced rise in demand will run up against supply constraints, thereby fuelling inflation, and rise in interest rates. This tightening of monetary policy, at a time when the economy needs accommodatory monetary policy, will end up crowding out private investments and off-setting the aggregate demand gains due to higher government spending. In contrast, monetary policy packs a much greater punch as an economic stabilization policy instrument. However, when there is a deep economic downturn coupled with interest rates touching the zero-bound, fiscal policy assumes a different character.

As Summers and DeLong write, there are atleast three distinguishing features of the current economic situation in many developed countries that leaves monetary policy without much traction and makes discretionary fiscal policy critical.

1. The absence of supply constraints and interest behavior associated with an economy constrained by the zero-bound means that the multiplier associated with fiscal expansion is likely to be substantially greater and longer lasting. The expectations of growth returning and raising inflation, and thereby lowering real interest rates, magnifies the multiplier.

2. Even very modest hysteresis effects through which output shortfalls affect the economy's future potential have a substantial effect on estimates of the impact of expansionary fiscal policies on future debt burdens. They find evidence that mitigating protracted output losses like those suffered by the United States in recent years raises potential future output. In other words, downturns have the potential to permanently lower the potential output and the trend rate of growth - "Large recessions may create labor-market as well as capital-stock hysteresis".

For example, the longer the economy stays depressed, the more likely that workers will quit the labour force altogether. Therefore, by putting these people back to work today, stimulus generates higher taxes not just this year but for years to come, lowering the long-term debt burden.

3. Extraordinarily low levels of real interest rates raise questions about the efficacy of monetary policy as a source of stimulus, and reduce the cost of fiscal stimulus.

In this context, Paul Krugman has this nice scatterplot of the changes in GDP growth rates against the change in government consumption among Eurozone economies. The correlation is unmistakably salient.

The "long" fiscal stimulus - a belated recognition of infrastructure spending?

Prof Tyler Cowen has a strange post. He points to the inadequacy of short-term stimulus spending, however large, to tide over deleveraging balance sheet recessions. He is therefore surprised that

"For all the talk of a 'large stimulus', you don’t hear much about a 'longer stimulus'."

He has this concern about short-term pump priming, whatever its size,

"The problem with a 'too small' stimulus is that you get an initial economic boost, but when the stimulus expires the economy slumps back down, as indeed happened in mid 2011. Ideally a stimulus employs some idle labor, stops it from depreciating, and tides those workers over until they can look for other jobs in fundamentally better economic conditions... If conditions are not improving soon, the ability of the stimulus to 'buy time' for those workers isn’t worth much... We end up having spent a lot of money to postpone our adjustment problems, rather than achieving takeoff. Deleveraging recessions last a long time, as shown by Rogoff and Reinhart. The need for continuing deleveraging implies that even a stimulus twice the size of ARRA won’t turn the tide."

In the circumstances, his suggestion,

"In those cases a well-designed stimulus program should not be so 'timely'. For a given presented expected value sum spent on stimulus, it is better to spread it out across the years. It is better to help a smaller set of workers for five years (or however many years it takes for most of the deleveraging to end), after which they are reemployable, than to temporarily boost a larger number of workers for two years, and then leave them back in the dust because deleveraging is still going on."

Here we go! There appears to be, to put it very charitably, an element of selective amnesia in Tyler's post here. This is effectively an admission of ideological failure (or, is it an error of judgement?) and an advocacy for focusing stimulus spending on creating durable public infrastructure assets atleast now.

In some sense, it is a classic case of the two-handed economist at work, albeit with a time lag in the action of the two hands. The one hand which had considered, debated and opposed the same when the ARRA was being formulated now appears to have changed track and embraced infrastructure spending when the earlier assumptions were proved wrong.

As early as late 2008, when the ARRA was being conceptualized, there was an intense and often acrimonious debate about the nature of the stimulus. Conservatives, who even then opposed any fiscal action, were willing to go only as far as tax cuts. They had opposed it on the grounds that there was no shelf of "shovel ready" infrastructure projects and that such spending takes time before its shows any stimulus effect on the economy. Marginal Revolution itself had directly posted and linked to several such views. In contrast, liberal economists like Paul Krugman, Mark Thoma and Brad DeLong felt that the recession was likely to persist for long and therefore preferred direct spending in infrastructure assets.

From hindsight, even the most conservative of economists would admit that the best course of fiscal policy action in late 2008 would have been to spend money on public infrastructure creation. The ultra-low interest rates, now certain to persist well into 2013 and atleast for a couple of years beyond that, would have provided unbelievably cheap financing for atleast 7-8 years. If in 2006, Congressmen and academics had been offered the prospect of accessing interest free loan for 8-10 years to repair America's battered infrastructure, many of them would have readily grabbed that opportunity. In fact, even the China-bashing Americans would have derived some vicarious pleasure from the realization that China was subsidizing America's infrastructure creation by offering virtually interest free loans!

In fact, Tyler's invocation of Reinhart-Rogoff now to fortify his argument about the pernicious nature of deleveraging recessions, appears to be a case of "what is sauce for the goose (is not) sauce for the gander"! Interestingly, Messers Krugman and Co had then invoked precisely the same duo to base their claim for infrastructure spending based stimulus. They had argued, based on the substantial body of empirical evidence presented by Reinhart-Rogoff about the average lengths of banking crisis induced recessions, that the Great Recession was likely to be a long drawn out one and therefore there was enough time for infrastructure spending to be effective.

The ideal course of action in late 2008 would have been to adopt a two-pronged approach, one which many of the aforementioned liberal/Keynesian economists did advocate, involving long-term stimulus on infrastructure creation and automatic stabilizers like unemployment insurance and food stamps to cushion the worst hit by the recession. Any tax cuts and other stimulus spending would have been an additional bonus.

I am inclined to believe that the impact of such spending on the economy as a whole would have been positive in many dimensions. Apart from the fact that it would have repaired or replaced the country's battered infrastructure, it would also have generated a significant multiplier on the economy on many fronts. It would have brought to work idle resources, encouraged businesses to not postpone investments, spurred market confidence (yes, the "confidence fairy"!) in the long-term health of the country, and so on. Given the fact that these investments were in any case necessary, one would also have to add the opportunity cost benefits of the ultra-low interest rates to calculate the multiplier.

In view of all the aforementioned, the final paragraph to Tyler's post is a sad commentary of the dark age of macroeconomics,

"Oddly, there is not much discussion about the length of fiscal stimulus. But there should be."

PS: I just did not have to energy to mine the numerous links in MR, and Krugman, Thoma and DeLong's blogs that contain the specific material from 2008-09 that I have alluded to in the post. I guess I am lazy! Anyways, interested readers could do so from here and here.

Automatic fiscal stabilizers and counter-cyclical fiscal policy

I have blogged extensively about the utility of fiscal policy in combating aggregate demand slumps, especially when the economy is facing the zero-bound in nominal interest rates.

However, unlike the more rules-based monetary policy, fiscal policy is subjective and deeply political. The classic fiscal policy alternatives like direct government spending on infrastructure face the problem of implementation lags. In contrast, automatic stabilizers kick-in immediately, being targeted on those most likely to spend any money provided to them. It no surprise that automatic stabilizers - unemployment insurance, food stamps etc - have among the highest fiscal multipliers.

The WSJ points to the apparent success of Sweden in managing its recovery from the Great Recession and attributes it to successful expansionary policies by both the government and the Riksbank. The Swedish economy grew 5.5% in 2010 and unemployment rate has fallen from its peak of 9% to 7%. Instead of high-profile direct spending and tax cuts, the Swedish government responded swiftly with automatic stabilizers to provide income, health care and other services to people who are unemployed. The Riksbank, initially lowered rates aggressively to zero, even taking it to minus 0.25% (savers had to pay 0.25% for the privilege of keeping deposits). Its quantitative easing program was more expansionary than even the Fed - the Riksbank's balance sheet was more than 25% of GDP in the summer of 2009, compared to 15% for the Fed.

Further, unlike many other developed economies, Sweden entered the recession in excellent fiscal health - its budget had a 3.6% of GDP surplus in 2007, to 3% deficit in the US. This gave the government enough cushion to indulge in extended fiscal expansion when recession struck. This was a result of a strong commitment, borne out of the bitter experience of its banking and economic crisis in early 1990s, to maintain a counter-cyclical fiscal policy.

Clice Crook points to the example of the US, where though the Obama administration came up with a large fiscal stimulus in 2009, mostly with tax cuts and direct spending, its impact was offset by the severe fiscal tightening by the local governments. He also writes about the relative lack of influence of fiscal stabilizers in the US,

"Two factors weaken automatic stabilizers in the US. First, the government is small, so economic fluctuations, other things being equal, move fiscal quantities less. Second, states are subject to balanced-budget rules. Much of the US government has to follow a pro-cyclical fiscal policy – cutting spending and raising taxes – during a recession."

The acrimonious debates surrounding fiscal expansion in the US underlines the need for a much greater role for automatic fiscal stabilizers. However, it is also important that these automatic stabilizers have automatic sunset clauses that ensure exit from fiscal expansion when the economy recovers. Mark Thoma makes an excellent case for greater use of automatic fiscal stabilizers during recessions.

In this context, Jeffrey Frankel, Carlos A. Vegh, and Guillermo Vuletin (pdf here) examined long-term fiscal policy in 94 countries (73 developing and 21 developed countries) over the 1960-2009 period and found that "the cyclicality of a country’s fiscal policy – a sign of its riskiness – is inversely correlated with the quality of the country’s institutions".

They examined the correlation between government spending and GDP for these countries over two periods, 1960-1999 and 1999-2009, and found a significant increase in countries with negative correlation (or counter-cyclical spending) over the two periods. In fact, among developing countries, those following counter-cyclical policies increased four-fold to 35% over the two periods. The graphic below indicates the correlation between spending and GDP for these countries in the 2000-09 period, with yellow and black bars representing developing and developed countries respectively.



The increase in counter-cyclicality in the conduct of fiscal policy by developing countries is evidence of greater maturity by policy makers and policy institutionalization in these countries. This maturity is corroborated by other indicators like reduced debt-to-GDP ratios in many developing countries. The authors "find that the cyclicality of a country’s fiscal policy is inversely correlated with the country’s institutional quality which includes measures of law and order, bureaucracy quality, corruption, and other risks to investment". They highlight the success of Chile with counter-cyclical fiscal policy and attributes it to institutional strengthening reforms since 1980s.

Prof Mankiw's "convenient agnosticism"!

Greg Mankiw is one of my favorite economists. His hugely popular text book is arguably the second-best window to learning the principles of Economics. His superstar blog has surely played a major role in enriching the debate on many political economy issues.

But his recent post explaining his ambivalence and agnosticism on extending unemployment insurance dismays me. In fact, it comes across as specious and positively disingenious, bordering on misleading and mis-directing the debate. Here is why

1. For a start, there is now enough literature on the pros (reduces household income uncertainty and props up aggregate demand) and cons (budgetary costs and lowers job search costs) raised by Prof Mankiw.

There is now ample evidence that extending the duration of UI does little to lower the marginal incentive to search for re-employment. Alan Krueger and Andreas Mueller have found that the time devoted to job search is fairly constant regardless of unemployment duration for those who are ineligible for UI.

Further, a recent study by the San Francisco Fed found negligible impact of UI extension on increase in unemployment rate - about 0.4 percentage point of the nearly 6 percentage point increase in the national unemployment rate since 2008.

2. The cases for and against any policy instrument varies depending on the broader macroeconomic environment. There cannot be a single applicable-for-all-conditions reasoning in favor or against a policy.

Consider this. The US economy is still recovering very slowly from a deep recession. Unemployment rates are on the rise and private sector job creation remains anemic. Inflation is low, aggregate demand is weak. Household balance sheets remain battered and will require more repairs. Monetary accommodation may have limited further traction. There is limited fiscal space available.

In the circumstances, doing nothing for lack of compelling quantitative evidence (of the success of the proposed intervention) and thereby letting the economy continue its present course has several dangers. The dismal macroeconomic conditions and prospects mean that the economy risks falling into a deep recession, from which recovery will be even more painful and long-drawn out. In fact, without recovery taking hold (and there is no evidence of this happening), the shares of debts and deficits will continue to grow.

The limited fiscal space available means that any stimulus spending should be directed at areas delivering the greatest bang for the buck. Money should be delivered to the hands of people who are likely to spend it immediately. A number of studies clearly indicate that among the various stimulus options, UI is among those with the largest multiplier.

3. The argument against any action on the grounds of conclusive enough evidence is all the more misleading since the inherent nature of an economy precludes any conclusive enough evidence on any reasonably complex economic policy intervention. Do we invoke this excuse to let a system take its own course, even when it is hurtling down into an abyss? Or do we weigh the relative probabilities of possible policy options and respond with the most cost-effective and most-likely-to-succeed option?

No one disputes that budget deficits and public debts should be lowered. The relevant question is whether this is the time to do so? What are the costs of contraction at this point in time? What is the cost of not doing anything and letting UI expire? Is there enough evidence in favor of the stimulative impact of UI?

In other words, the choice is between risking a deflationary recession with rising shares of debts and deficits and untold pain for the vast majority of people, and stimulating a recovery by running up short-term deficits. Reasonable people would choose the later as the lesser evil.

4.

"So when I hear economists advocate the extension of UI to 99 weeks, I am tempted to ask, would you also favor a further extension to 199 weeks, or 299 weeks, or 1099 weeks? If 99 weeks is better than 26 weeks, but 199 is too much, how do you know?"

This too has echoes of the first point. Prof Mankiw surely knows that there cannot be a not-too-high, not-too-low, and just right magic figure for the duration of UI applicable for all situations. In fact, the issue to be discussed here is not whether UI duration is too-high or too-low, but the benefits of its extension as opposed to costs. At issue is not what constitutes the optimal duration of UI, but whether fiscal stimulus is required and what is the most effective form of stimulus. If we assume that the moral hazard of UI is negligible and that we need fiscal stimulus, then the case in favor of extending UI is pretty strong.

5.

"It is also conceivable that the amount of UI offered in normal times is higher than optimal and that a further extension would move us farther from what is desirable."

On this, there is already enough evidence that US stands on the negative side of the optimum UI duration scale to Europe's positive side. So the argument on the possibility of deviating farther away from the desirable may be superfluous.

6. Prof Mankiw's argument is based on lack of "compelling quantitative" evidence on the optimum period of UI. He gives the impression that he is unable (or unwilling) to take a decision on UI till he has "compelling quantitative analysis" about "how generous the optimal system would be". Unfortunately, by setting his conditions in so comprehensive a manner, he has virtually ensured that he will never need to take a decision to support extension of UI. In any case, how much evidence is "compelling" enough?

He knows all too well that it is impossible to conclusively quantify the optimal duration of UI - not now, never in future too. In the circumtances, this logic is a convenient excuse for not doing anything. This naturally translates into letting the UI benefits expire. In other words, you vote against extension without appearing to directly oppose it!

And such reasoning has now become the convenient excuse for conservative economists to recuse themselves from supporting government interventions in many areas. The pathological opposition to tax increases too works on similar lines. Once again, it is impossible to conclusively prove whether the system falls on the upward (left) or downward (right) side of the Laffer curve.

Similarly, conservatives point to the persistent high unemployment rates and weak economic growth and argue that the fiscal stimulus measures failed. The difficulty of estimating the counterfactual condition (without stimulus) means that an argument set on such terms of reference cannot be easily refuted.

Taken to its extremes, such reasoning can be used to justify almost any position. In social sciences like economics, it is impossible to conclusively quantitatively prove that a proposed intervention or policy is required. Decisions on whether fiscal stimulus is necessary and what types of spending is required, will never be proved in a universally compelling manner.

Tax cuts have the lowest fiscal multiplier

The latest CBO assessment of the impact of ARRA once again (here is the earlier evidence) highlights that tax cuts are among the least effective of fiscal stimulus measures. This assumes significance in view of the decision pending before the US Congress on whether to extend the Bush tax cuts beyond its expiration period at the end of the year.

In fact, even the higher estimates on tax cuts on higher income people and corporates, revealed fiscal multipliers less than unity. In contrast, direct government spending, transfers to states, local governments, and individuals ((such as increased food stamp benefits and additional weeks of unemployment benefits), yielded higher multipliers.




In an excellent recent post, Mark Thoma examined the evidence on the impact of the Bush-era tax cuts of 2001 and 2003. The Bush tax cuts were done in two phases - in 2001 the top statutory income tax rate was reduced from 39.6% to 33%, while in 2003 the tax rate on both capital gains and dividends was lowered to 15%.

I have blogged earlier that far from having any positive impact, the marginal income tax rate increases of 2001 adversely affected household incomes, unemployment rates, and deficits. There is a large body of evidence to show that capital gains and dividend tax cuts had little or no impact on stock prices and corporate payouts. Further, the tax cuts on capital gains and dividends has also been held responsible for the massive widening of income inequality in the US in recent years.

Fiscal multipliers for tax cuts and spending

I have blogged here and here about fiscal multipliers for different policy choices. Among the different fiscal stimulus spending choices, there is heated debate on the relative effectiveness (in terms of its bang for the buck or fiscal multiplier) of tax cuts and direct spending.

In the US, with the controversial Bush original tax cut in place since June 2001, set to expire by end of this year, there is an intense debate raging about its future. Republicans expectedly want them to continue and made permanent, whereas Democrats are broadly converging to the view that it should expire for those with incomes beyond $250,000.

Mark Zandi has this revised and latest examinaton of the fiscal multipliers of various stimulus measures.



As can be seen, direct government spending - through unemployment benefits, food stamps, work sharing or infrastructure spending - top the list, giving more than a dollar's worth of stimulus for a dollar's worth of spending, while cuts to taxes affecting businesses and upper-income individuals - such as the corporate, dividend, capital gains and alternative minimum taxes - gives much less.

However, since tax cuts are politically easier to push through, tax rebates aimed at those likely to spend (as against save it), jobs tax credit to spur investment and hiring, and cuts in payroll tax (which finances Social Security and Medicare and is paid by both businesses and workers) may be the most effective.

See also this paper by Alan Auerbach and Yuriy Gorodnichenko who find that "large differences in the size of fiscal multipliers in recessions and expansions with fiscal policy being considerably more effective in recessions than in expansions" and that "controlling for predictable components of fiscal shocks tends to increase the size of the multipliers".

Update 1 (25/9/2010)

Analysis show that the 2008 Bush tax cuts failed.

The case for more stimulus

This is a long post... a sort of summary of the debate surrounding continuation of expansionary policies. With the US economy getting better much slower than required (especially to get normalcy restored in the job market), the debate about further expansionary stimulus measures has picked up.

The deficit hawks point to the burgeoning public debts and raise the spectre of sovereign defaults (made more salient by the problems facing the PIIGS in Europe) and unhinged inflationary expectations. They call for an immediate end to all stimulus spending and initiation of measures to rein back the deficits through spending cuts and interest rate increases to pre-empt inflation.

On the other side are those who point to the bitter decade-long experience of Japan with deflation and economic stagnation in the nineties, and caution against any premature exit from the expansionary stimulus policies of the last two years. They worry that the household and business balance sheets are so badly damaged and the unemployment rates too high that any fears of "crowding-out" of private investments and inflation are unfounded. They advocate continuation of the expansionary fiscal policies to boost aggregate demand and create jobs (and also prevent lay-offs) and further loosening of monetary policy to keep credit flowing, longer-term interest rates low, and even stoke some inflation. They point to the declining trend in prices and stable long-term bond yields to justify their claim.

However, the debate on the relative effectiveness of fiscal and monetary policies, continues to rage unabated. I have already blogged about the problems faced by monetary policy when facing the zero-bound in nominal interest rates. Economists like Joseph Gagnon, advocate further monetary easing in Japan, euro-zone and US through large purchases of long-term bonds (pdf here) by the central banks to reduce long-term interest rates. He points to the extremely anemic economic environment and the very low inflation rates in these countries to argue for such expansionary policies.

Unlike normal times, increasing the monetary base when faced with the zero-bound in nominal interest rates will have an expansionary impact only if "people believe it signals higher inflation later". However, as Mark Thoma writes, the Fed's carefully constructed, stellar inflation fighting reputation raises serious doubts on its credibility to commit itself to future inflation. People are most likely to think that the Fed will pull the levers at the first signs of inflation gathering steam. Politically too, the paranoia about inflation, active even when deflation is staring us, cannot be avoided when prices start to rise.

All this means that fiscal policy becomes the preferred option, especially when interest rates have lost traction. However, as Mark Thoma explains, during normal times, with both full-price adjustment classical models and sticky-price New Keynesian models and an inflation targeting central bank, fiscal policy generates multipliers less than one,

"When government spending goes up during normal times, inflation increases, and sharp increases in the real interest rate are needed to return inflation to its target value. The sharp increase in the real interest rate offsets the increase in output brought about by the increase in government spending, and this is what makes the multiplier small in this case. Under reasonable parametrizations, there "really isn’t much fiscal policy can do."

More particularly, with strict inflation targeting, the multiplier is less than one. When the Fed follows a Taylor rule instead of strict inflation targeting, the multiplier is larger, but still less than one (though not always, it could even be less than the multiplier for strict inflation targeting under some conditions). If monetary policy maintains a constant real rate instead of following a Taylor rule, the multiplier is equal to one. This means that during normal times, sticky price models predict fiscal policy multipliers of a magnitude less than or equal to one, with the exact magnitude depending upon the rule the Fed follows, i.e. how the real interest rate responds to fiscal policy changes."

However, as Micheal Woodford and others have found, things change when the interest rate is touching the zero-bound. As Mark Thoma writes, this happens because the expectations of increase in fiscal spending raises inflationary expectations and thereby lowers real interest rates,

"In general, at the zero bound fiscal policy multipliers are greater than one, and this remains true under strict inflation targeting... The larger multiplier occurs because the increase in government spending increases inflation (more precisely it reduces the rate of deflation). If the crisis is expected to last another period with some probability, as it will in the model, then government spending is expected to persist as well and expected inflation will rise. The increase in expected inflation lowers the real interest rate when the zero bound is a constraint (even with strict inflation targeting), and the lower real interest rate generates additional economic activity.

Note that the source of the increase in expected inflation is the expected increase in government spending in the next time period. All that's required for expected inflation to rise is that fiscal policy is expected to persist another period. However, the Fed won't do anything in response to the rise in inflation expectations because under the assumptions of the model the target interest rate remains negative."

The case in favor of fiscal policy (over monetary policy) is explored in detail with numerous links here and here. The IMF too recently examined the impacts of fiscal policy across economies, under various conditions, and came to favorable conclusions about the effectiveness of fiscal stimulus spending policies. The IMF's latest fiscal monitor too captures the favorable impact of the stimuluses.

Among fiscal policy options tax cuts and direct spending have been amongst the most popular. This is despite the fact that welfare spending through automatic stabilizers like food stamps, unemployment insurance, and nutritional support for children, may be more effective in containing the most debilitating effects of a recession. As the length of the slowdown increases and unemployment rates remain stubbornly high, assistance to state and local governments are fast emerging as an important source of fiscal policy intervention.

Mark Thoma points here and here to the problem posed by the deteriorating fiscal positions of state and local governments in the US that reflects in the rising job losses in those areas. The expansionary impact of federal fiscal stimulus is being countervailed by the contractionary impact of spending cuts by state and local governments.



In the search for swift-acting fiscal stimulus interventions with large multipliers (and socio-economic impact), assistance to state and local governments would surely be amongst the most effective.

And as Brad De Long writes, when faced with high unemployment rates, weak investment and spending environment, and anemic growth expectations, short-run deficits may be more expansionary and beneficial in the long run and belt tightening contractionary and harmful. In these times, he calls for prescriptions suited for "depression economics", wherein the beneficial effects of government spending and tax cuts will more than off-set the harmful effects of increased debt burden.

He compares the arithmetic of the relative costs of fiscal expansions during normal times and during the present times (where rules of "depression economics" applies) and finds that while "expansionary deficit-boosting fiscal policy is simply a non-starter in normal times", it generates "more income and employment now... in return for sacrificing only a tiny bit of production each year in the future, when we believe that we will be richer and will mind the reduction significantly less".

This is because, unlike normal times, more government spending now will not lead the Federal Reserve to raise interest rates to fight inflation, there is no "crowding out", boost to production (from a spending program) creates a substantial reflow in taxes that makes the spending program a bargain, and the government can borrow at "uniquely favorable terms" and thereby keep debt serrvice burdens manageable. His conclusion

"Each dollar of missing production and each unemployed worker right now is much, much more painful to the country and a much greater loss to human welfare than a dollar of missed production and an unemployed worker in normal times."

Tyler Cowen writes that reduction in real interest rates will not make much of a difference in investment decisions of businesses since the investment determining constraint is the "hurdle rate", which does not change by much despite the recession. He points to the fact that the "hurdle rate" in investments is in the range of 20-30% (to account for agency problems and related transaction costs) and therefore small reductions in interest rates have limited impact. In Econ 101 terms, real interest rates become a largely non-binding constraint since the elasticity of new projects to changes in real interest rates is very low (at the prevailing hurdle rates).

However, a logical extension of this line of arguement would mean that interest rates are always a non-binding constraint on investment decisions. Since the "hurdle rates" are in the range of 20-30% even during normal times, any small interest rate changes (and rate changes will always be small, a few tens of basis points, when compared to the "hurdle rate") will have no impact on the investment decision.

In another post, he also feels that the real issue is lack of trust, which has forced businesses to under-invest and households to under-spend. This lack of trust causes consumers, companies and financial firms to be more cautious than they’d otherwise be, and results in sub-par growth and occasional market frights. This is what economists like Robert Shiller have been pointing to for some time now, and what Keynes himself alluded to when he referred to the depressing role of "animal spirits".

And since the challenge is to get the "animal spirits" active and market-confidence restored, as is increasingly becoming evident (especially when faced with the zero-bound), aggressive fiscal policy interventions are required, irrespective of the what the short-run deficit. The alternative to this is the strong possibility of slipping ever deeper into recession and thereby increasing the costs (and deepening the fiscal strains) of recovery.

See also David Leonhardt here and here.

Why tax cuts are still not the remedy?

Nomura economist, Richard Koo had argued that currently we are passing through a "balance sheet deflation", wherein businesses and households have battered balance sheets which can be repaired only with fiscal policy (especially when monetary policy has lost traction). The magnitude of this cycle has been amplified by the fact that majority of the balance sheets were inflated by the valuations of assets which were in the first instancee purchased with unsustainable levels of debt. Once the bubble burst and the balance sheets exploded, forcing margin calls, defaults started, thereby amplifying the problem manifold.

Mark Thoma argues that in "balance sheet recessions" like the current one - where household (and business) balance sheets have been devastated by plunging asset (equity and homes) values - tax cuts can play an important role in repairing household balance sheets. This assumes importance in view of the fact that household spending, which is fundamental towards boosting aggregate demand, will get back to normal only when the debt holes in household balance sheets are refilled.

Ultimately, as Econ 101 teaches us, economic growth has to come by way of increased aggregate demand. For this to happen, repaired balance sheets help in so far as it encourages consumers to spend and businesses to invest. However, this indirect approach comes up against some pitfalls.

Taking households alone, the two major sources of balance sheet crisis are those arising from plunging asset values (and resultant "income-loss effect" which stunts consumption expenditure and causes a rise in effective mortgage values) and unemployment generated through lay-offs.

In the present case, the damage inflicted on balance sheets through the sub-prime meltdown and declines in asset values are too large, a few trillions of dollars, to be compensated with a few billions of dollars of tax cuts. For sure, these tax cuts, if properly targeted, will go into repairing balance sheets. But its impact is not likely to be much, unless it is relying on a vain hope that asset prices and market confidence will rebound adequately and in quick time and thereby repair the balance sheets. In the circumstances, one-off tax cuts, like the Bush tax rebate of Spring of 2008, will end up getting saved (and/or repay debts) and making limted impact on consumption.



Further, the damage caused by the largest post-war increase in unemployment rate cannot be repaired in any meaningful manner with one-off tax cuts. It can be argued that unemployment insurance is itself a form of targeted monthly/weekly income tax credit. However, as economists like Paul Krugman have been arguing, it is impossible to provide any significant boost to aggregate demand on a sustainable basis without quickly bringing down the record unemployment rates. Unfortunately, as Brad DeLong recently pointed out, unemployment appears far removed from the concerns of policymakers in the US.

It is in this context that the role of the multiplier assumes significance. Assuming a large enough multiplier for direct government spending, especially during recessions with the macroeconomic environment like now (and there is a very large and growing body of research to add credence to this claim), the arguement against tax cuts gains strength.

The other important objection against supply-side policies like tax cuts comes from the current macroeconomic environment. Despite widely expressed fears of impending inflationary spiral, all major indicators appears to inform that it is deflation and not inflation that should be the cause for concern for policymakers.

A cursory reading of the classic AS-AD curve indicates that when the supply increases without a commensurate increase in the AD (and from the aforementioned evidences, especially at a time like now, it is most likely that a major share of the tax cuts will not go into increasing AD), the prices will fall. In other words, tax cuts in a depressed economic environment like now, are likely to generate deflationary pressures.



In the final analysis, given the precarious fiscal position in most advanced economies, the priority should be on funneling resources into policies that deliver the greatest bang for the buck. In an ideal world, where resources are plentiful, all instruments of fiscal policy - tax cuts, infrastructure spending, and welfare measures - should be deployed.

However, when faced with a macroeconomic environment like now, policies than can directly boost the aggregate demand and generate plus-one multiplier are surely more attractive than ones that work their way slowly by repairing household and firm balance sheets. Moreover, policy makers should hope that they get a helping hand in repairing the balance sheets from the continuing expansionary monetary policy and a resurgence in the asset markets.

More on impact of fiscal stimulus

Latest addition to the fiscal stimulus impact literature comes from an excellent survey by the IMF.

The paper studies the short-run economic impact (or size of fiscal multipliers) of temporary government fiscal policy actions in lessening the depth and duration of the slowdown, using seven commonly used structural models of national economies and global economy. In view of the fact that these models are designed for normal situations when central banks undertake monetary contraction in the aftermath of a fiscal expansion to pre-empt inflationary pressures, the authors explore an extra dimension by examining the impact of fiscal stimulus when the central banks follow a policy of monetary accommodation. They conclude,

"There is no such thing as a simple fiscal multiplier. The size of the response of the economy to temporary discretionary fiscal stimulus depends on a number of factors, including most importantly the type of fiscal instrument used and the extent of monetary accommodation of the higher inflation generated by the stimulus. Temporary expansionary fiscal actions are most effective when the fiscal instrument is spending or well-targeted transfers, and when in addition monetary policy is accommodative. On the other hand, permanent stimulus, that is a permanent increase in deficits, is much more problematic than temporary stimulus. It leads to a long-run contraction in output, but in addition it substantially reduces short-run fiscal multipliers. Finally, the G20 stimulus should have significant effects on global GDP in 2009 and 2010."

The graphics below captures the relative impacts of various fiscal interventions in the US and Europe (as estimated by different ageencies/models) over time, with varying years of monetary accomodation.

1. Impact of targeted transfers (is very effective with extended monetary accommodation)





2. Impact of labor income tax (has minimal impact)





3. Impact of corporate income tax (has the least impact, even with monetary accommodation)



4. Fiscal stimulus impact on real GDP across the world

Fiscal multipliers under different conditions

I have blogged here, here, here, and here about the debate on the impact of fiscal stimuluses, more specifically the fiscal multipliers associated with different types of stimulus spending. While conservatives have argued that multipliers are essentially zero and therefore find no merit in fiscal spending, supporters point to substantial multipliers in government spending, especially in deep recessions when demand is weak and business investments have dried up.

A Vox post by Ethan Ilzetzki, Enrique G. Mendoza, and Carlos A. Vegh (full paper here) adds a new dimension to the debate by claiming that "fiscal multipliers are much weaker in countries that have high debt, lower income, flexible exchange rates, and greater international openness". They also find that preduicting fiscal multilpiers with any degree of certainty is even more difficult for developing economies. Their findings are

1. The response of output to increases in government spending is smaller on impact and considerably less persistent in developing countries than in high-income countries.
2. Fiscal multipliers are much larger in economies operating under predetermined exchange rate regimes than under flexible exchange rates.
3. Relatively closed economies have much larger multipliers than relatively open economies.
4. The output response to increases in government spending is short-lived and much less persistent in highly indebted countries than in countries with a low debt to GDP ratio.
5. The multipliers for the US in the post-1980 period are small both in the short and long-run. On the other hand, multipliers for government investment are large.

These findings carry important policy implications. They lend weight to the need for globally co-ordinated stimulus spending policies, among atleast the major economies, in an increasingly inter-connected global economy, failing which protectionist backlash is inveitable and some degree of protectionism is even desirable. With fiscal expansions having considerable positive externalities, a substantial fraction of the stimulus spending leaks out to the rest of the world through higher imports etc.

It also supports the important but always-forgotten holy grail in fiscal policy making - follow a counter-cyclical fiscal policy and build up surpluses during good times and unwind them and run up deficits during downturns. Governments, especially in the developing world, who have tended to follow either the populist "spend more when the going is good" or the a-cyclical business cycle neutral tax and spending policies, are left with limited fiscal space when the bad times arrive. India is a case in point.

The authors find evidence of "crowding out" effect in developing countries, where an additional dollar of government consumption crowds out some other component of GDP - investment, consumption, or net exports - in the long run. However, this finding may actually turn out to be the opposite in deep recessions, when household demand and private investments become frozen, and government spending can play the important role of "crowding in" aggregate demand. During the current recession, most of the emerging economies did not experience the same extent of output contraction as the developed economies, and therefore fiscal expansions in these countries may not have supplied the same boost to aggregate demand as in the latter.

Mostly Economics points to the US CEA's latest impact assessment of the $787 bn ARRA stimulus spending plan. It finds that the stimulus spending changed the trajectory of the economy toward moderating output decline and job loss; it added roughly 2.3 percentage points to real GDP growth in the second quarter and is likely to add even more to growth in the third quarter; caused employment in August to be slightly more than 1 million jobs higher than it otherwise would have been; and it added between 2 and 3 percentage points to baseline real GDP growth in the second quarter of 2009 and around 3 percentage points in the third quarter. It estimates a very high fiscal multiplier of 2.3 in Q2 2009 and 2.7 in Q3 2009. It also finds that assistance to states played a critical role in helping states facing large budget shortfalls because of the recession by increasing employment relative to what would have happened without stimulus.

Mark Thoma, as always, captures the debate on stimulus multipliers here. Paul Krugman has this response to Robert Barro's assertion of a multplier less than one. The Economist has a nice summary of the multiplier debate and explains why it is so difficult to make any predictions about them given the wide variations in economic conditions.

A recent NBER working paper by Lawrence Christiano, Martin Eichenbaum, and Sergio Rebelo argues that "the government-spending multiplier can be much larger than one when the nominal interest rate does not respond to an increase in government spending". They claim that if the nominal interest rate is governed by a Taylor rule, it rises in response to an expansionary fiscal policy shock that puts upward pressure on output and inflation, and thereby renders the multiplier small. However, when nominal interest rates does not respond to an increase in government spending (when the zero lower bound on the nominal interest rate binds), their model finds that the multiplier is very large. Taking this model and its line of explanation, the effectiveness of fiscal spending is likely to be limited for many developing countries, including India, where the nominal rates are high and where inflationary pressures makes interest rates more sensitive to revisions.

And their conclusion has great relevance for the major developed economies which have nominal interest rates kissing the zero-bound, "In such economies it can be socially optimal to substantially raise government spending in response to shocks that make the zero lower bound on the nominal interest rate binding... for government spending to be a powerful weapon in combating output losses associated with the zero bound state, it is critical that the bulk of the spending come on line when the lower bound is actually binding."

They argue that when the economy is touching the zero-bound and the output falls, a deflationary spiral is unleashed that drives up the real interest rates, which in turn leads to an increase in the level of desired savings. Since investment is zero during such recessions, the aggregate saving must be zero in equilibrium, and the total fall in output required to reduce desired saving to zero is very large. And about how fiscal spending works in a recession when this zero bound is binding, they write,

"This (spending) increase leads to a rise in output, marginal cost and expected inflation. With the nominal interest rate stuck at zero, the rise in expected inflation drives down the real interest rate which drives up private spending. This rise in spending leads to a further rise in output, marginal cost, and expected inflation and a further decline in the real interest rate. The net result is a large rise in inflation and output. In effect, the increase in government consumption unleashes an inflationary spiral that counteracts the deflationary spiral associated with the zero bound state."

Update 1 (28/8/2010)
Mark Zandi (via Ezra Klein) has this graphic which examines the bang for the buck for various types of stimulus spending in the US.

More on tax cuts Vs government spending

I had blogged earlier about the debate about the relative merits of tax cuts and government spending as fiscal stimulus measures. This blog has consistently argued that direct government spending programs pack a much greater punch in stimulating aggregate demand than indirect ones like tax cuts.

An IMF working paper by Emanuele Baldacci, Sanjeev Gupta, and Carlos Mulas-Granados, which studied the effects of fiscal policy response in 118 episodes of systemic banking crisis in advanced and emerging market countries during 1980–2008, finds that "timely countercyclical fiscal measures contribute to shortening the length of crisis episodes by stimulating aggregate demand".

About the superiority of government consumption over government investments and tax cuts, they write,

"The composition of fiscal expansions matters for crisis length - a point that has not been studied in the literature. Stimulus packages that rely mostly on measures to support government consumption are more effective in shortening the crisis duration than those based on public investment. A 10 percentage point increase in the share of public consumption in the budget reduces the crisis length by three to four months. Reducing the share of income taxes is less effective than consumption taxes in shortening the length of a banking crisis."

They also draw attention to the policy trade-off between measures to stimulate short-run aggregate demand that impacts output and employment relatively fast, and delayed (public investment-driven) stimulus that has a larger impact on productivity and economic growth,

"The quality of the fiscal stimulus package matters most for post-crisis growth resumption, with fiscal responses relying largely on scaling up the share of public investment in the budget showing the largest positive effect on medium-term output growth. A one percent increase in the share of capital outlays in the budget raised post-crisis growth by about one-third of one percent per year. Income tax reductions are also associated with positive growth effects."

And of relevance to fiscally constrained countries like India,

"Initial fiscal conditions matter for fiscal performance during shocks. In countries with high pre-crisis ratios of public sector debt to GDP, lack of fiscal space not only constrains the government's ability to implement countercyclical policies, but also undermines the effectiveness of fiscal stimulus and the quality of fiscal performance. In countries with high debt, crises lasted almost one year longer. The effect of high public debt on duration completely offset the benefits of expansionary fiscal policies in these countries...

These findings point to the importance of creating fiscal space and enhancing macroeconomic stability in tranquil times to limit the risk of falling into crises and to enhance the effectiveness of policy responses when exogenous shocks hit countries... fiscal policy responses may not be effective when initial fiscal conditions are poor and fiscal space is limited. High public debt levels and past macroeconomic instability limit the scope for countercyclical deficit expansions and hamper the effectiveness of fiscal stimulus measures as markets perceive the higher future fiscal risks entailed by larger deficits"

Menzie Chinn points to an excellent IMF staff poistion note that summarizes the fiscal policy responses across the globe and makes projections about the fiscal positions of these countries. Among the G-20, it estimates fiscal deficits in both 2009 and 2010 to be 5.5% of GDP above their pre-crisis (2007) levels. Of this, crisis-related discretionary measures is estimated to be 2% of GDP in 2009 and 1.6% of GDP in 2010, with the rest of the change in fiscal balances reflecting primarily the automatic fiscal stabilizers and revenue losses associated with extraordinary declines in asset and commodity prices.




It finds government spending — either for consumption or investment — more effective than cutting taxes,

"While government spending results in a direct increase in aggregate demand, tax cuts might not be fully spent (although increased saving may have a beneficial impact over the medium term in repairing household balance sheets). The IMF’s Global Integrated Monetary and Fiscal Model (GIMF) yields low fiscal multipliers for cuts in labor taxes and lump-sum transfers (0.2–0.5); and high multipliers for government expenditure (1.6–3.9) and targeted transfers (0.5–1.7). Zandi (2008) finds larger fiscal multipliers for infrastructure spending and targeted transfers (1.7) than for general tax cuts (0.3). Finally, a 2003 UK Treasury study based on the European Commission’s QUEST model finds larger one-year fiscal multipliers for government spending (0.3–0.7) than for tax cuts (0–0.3). This said, as noted earlier, it often takes long to activate spending without wasting public resources, especially for new programs."

As Mark Thoma points out, tax cuts work in two ways. First, if they are spent and not saved or used to repay debts, this spending "stimulates aggregate demand, output, and employment". Second, if saved or used to repay off debts, it helps "refill damaged balance sheets" and thereby "shorten the length of recessions". He argues in favor of a portfolio of fiscal stimulus policies - government investment, government consumption, and tax cuts. I fully agree with his summary of the debate and his finding that though the US fiscal stimuluses contained more than enough measures devoted to long-run economic growth, it had far too few devoted to simulating aggregate demand immediately,

"Tax cuts on consumption and government consumption have a relatively immediate impact both on aggregate demand and on the rate at which balance sheets are repaired, and income tax cuts along with spending on infrastructure are better at enhancing long-run growth... tax cuts can help to shorten recessions as described above, and this effect occurs both because tax cuts help to repair balance sheets when the tax cuts are saved, and because they stimulate consumption. But the effectiveness of the tax cuts in the short-run could have been improved by targeting consumption rather than income, and government consumption may have had an even larger effect."

Gary Burtless highlights the achievements of the fiscal stimulus in the US that in the six quarters since the end of 2007 transferred more than $830 billion to Americans’ personal disposable income through personal tax payments and social insurance contributions. Transfer payments to households increased $382 billion, or 22%.

Update 1
Miguel Almunia, Agustín S. Bénétrix, Barry Eichengreen, Kevin H. O’Rourke, and Gisela Rua (full paper here) gathers data on growth, budgets and central bank policy rates for 27 countries covering the period 1925-39 and shows that where fiscal policy was tried, it was effective. They find fiscal multipliers as large as 2 in the first year, before declining significantly in subsequent years.

In the absence of adequate awareness and knowledge about fiscal policy and with Central Banks too closely tied up with the Gold Standard, governments generally followed conservative policies to combat the Depression. Japan and Italy were exceptions.

They also find that contrary to widespread belief that monetary policy is ineffective in near-zero-interest-rate (liquidity trap) conditions, in the 1930s it accommodating monetary policy helped, by transforming deflationary expectations and by helping to mend broken banking systems.

Two handed economist - multiplier debates

The debate about the utility of various fiscal policy measures in combating economic recessions has revolved around the magnitude of the respective fiscal multipliers. These fiscal multipliers, which measures the impact of an increase in government spending on GDP and employment, are critical for determining the appropriate size and timing of countercyclical fiscal policy packages. In evaluating the recent $787 bn US fiscal stimulus, Christina Romer and Jared Bernstein had estimated the fiscal multiplier at 1.6.



Greg Mankiw points to two studies that sought to calculate the fiscal policy multipliers in a new Keynesian DSGE model when the economy is at the zero interest lower bound. On the one hand, Martin Eichenbaum, Lawrence Christiano, and Sergio Rebelo find large multipliers, while on the other hand, John Cogan, Tobias Cwik, John Taylor, and Volker Wieland arrive at far lower multipliers!

The first study finds that the government spending multiplier can be very large when the zero bound on nominal interest rates is binding. In contrast, the latter claim that the multipliers are less than one as investments and consumption gets crowded out. Brad De Long draws distinction between multipliers on spending during normal times and when the rates are touching the zero-bound. He argues that the Cogan-Cwik-Taylor-Wieland model is a model of a small multiplier in an economy away from the zero nominal interest rate bound when central banks are targeting inflation.

Update 1
Paul Krugman responds here and here.