Comment

Abstract

Previous articleNext article FreeCommentHarald UhligHarald UhligUniversity of Chicago, NBER, and CEPR Search for more articles by this author University of Chicago, NBER, and CEPRPDFPDF PLUSFull Text Add to favoritesDownload CitationTrack CitationsPermissionsReprints Share onFacebookTwitterLinked InRedditEmailQR Code SectionsMoreIntroductionThis paper is written as a contribution to the science of economics, and it is a beautiful contribution indeed. But let’s not mince words. While the paper does not make a policy recommendation per se, it nonetheless seems to me quite plausible that readers will cite that analysis as lending strong support for the perspective that Germany ought to rescue Greece and other euro-zone countries in economic troubles, per substantially increasing its own fiscal spending. Allow me to add some additional bits of analysis of my own. While I shall not make any policy recommendation per se, either, it may then be plausible that readers of both pieces end up more sceptical regarding that rescue recommendation than if they just read the fine paper by Blanchard, Erceg, and Lindé.Even while news has shifted to other topics, the Eurozone crisis of 2010 still lingers on. Indeed, and in many ways, we have not escaped it yet. Things are dire and far from well. Unemployment rates in Greece and Spain, say, are still ridiculously high. The Italian banking system is fragile. Good policy solutions are sorely needed. So, what to do? Are the Germany-proposed “austerity” measures the right approach? Or should one follow recommendations more in line with the views of the former Greek finance minister Varoufakis and plead for deficit spending and Keynesian multipliers? The simplest version of that route would be to have, say, Greece do lots of fiscal spending itself, bankrolled by Germany and other Eurozone members. Clearly, these proposals did not go far, politically.This paper makes a substantial contribution to this debate, using a variety of models and analyses to shed light on the key issues. As such, it is a sorely needed, surely welcome, thoughtful, and insightful addition. Allow me to put it that way. The policy solution considered here is a new variant of the more-fiscal-spending approach and argues that Germany should directly do that fiscal spending. Indeed, why reroute those resources via Greece? Sure, the Greek government would presumably spend most of its resources on Greek goods and services, where that spending would plausibly do the greatest amount of good for the Greek economy, but obviously, the German government could do just the same. In the language of the paper, and in section V.B, this is stated as follows: “Larger welfare improvements could be achieved by channeling relatively more of the fiscal stimulus to the periphery.” Sure, it is nice to see that the paper delivers this conclusion also formally. Perhaps the German government could choose a slightly different spending pattern. Any Germans out there in favor of free vacations in Greece, courtesy of the German government? Perhaps that would be easier to sell politically in Germany. But, overall, it still feels like the conventional recipe of doing more spending in Greece, bankrolled by the German government.So what, exactly, does the paper do? It is a very well-crafted exercise of examining fiscal stimulus at the zero lower bound in a two-country version of the well-researched, much-liked, but also much-maligned three-equation New Keynesian model. This is very mainstream these days, and these authors provide a nice addition. The analysis is complemented with a welfare analysis, with VAR evidence on monetary and fiscal shocks, and more a complete medium-scale New Keynesian model with capital, and so forth. I highly recommend it to any student or researcher, who wishes to read and learn a good benchmark model in that literature, and see how to extract results and insights. The paper is ambitious, thoughtful, informative, lucid, and full of interesting results and detail reflecting a large amount of work. Once one buys into the approach, one can only applaud the authors in what they deliver in this paper.Welfare ComparisonsThe paper is also remarkably honest. It provides a welfare evaluation of a particular policy proposal, first using an ad hoc welfare function. With that, the advantages of the fiscal expansion are quite substantial. Many authors might have stopped there, satisfied with the result. But the paper also provides an evaluation based on the full calculation of the discounted expected utility of the representative household in each member state. It provides a detailed and lucid discussion of the rather intricate channels of how the policy affects the welfare in the various countries. It finds in section V.B, for example, that “an expansion of core spending causes periphery welfare to deteriorate considerably” and admits that “these results differ dramatically from the implications of the ad hoc loss function in which the higher import content was a major plus for welfare in both CU members.” That is not the result the authors had hoped for, I guess. After looking at these sobering results from a number of angles (some of which help to hide or ignore some of the less benign impacts of these policies), the authors remarkably conclude that “our sense is that the welfare benefits … probably lie between the two measures, but tilt more in the direction suggested by the simple ad hoc loss function.” The authors are entitled to their assessment, of course. If you are convinced by that, wonderful. I personally feel more enamored by a model-consistent analysis including the welfare calculations. Otherwise, one has to wonder whether there is any model out there for which some ad hoc loss function would not deliver some given policy conclusion. I shall admit that theirs does not look entirely unreasonable, but still. The authors provide a careful discussion of these various measures and channels, and for readers like me, who are more inclined to trust results based on a consistent anlysis, it offers plenty of reasons to remain skeptical about the benefits of German fiscal spending to rescue Greece.Paradigm SkepticismNow, these skeptical conclusions are already available, conditional on the paradigm offered in the paper. Their paradigm is mainstream; it is in much use. But can one trust it enough to deliver sufficiently reliable policy advice? I believe that it is good to have it as one of the tools out there, that its conclusions and mechanisms are worth studying with sufficient care, but that any policy conclusions should be met with a considerable dose of skepticism. Allow me to elaborate my reasons.The Phillips CloudAt the heart of the New Keynesian approach is the Phillips curve and its relationship between inflation and the output gap, pronounced to be reliable enough as a basis for policy interventions. But how much of a Phillips curve is really there, in the data? The New Keynesian models typically use a version ofπt=βEt[πt+1]+κxt+εt,where πt is inflation from t −1 to t, xt is a measure of the output gap, κ is a coefficient, and ɛt is a disturbance term. The output gap is a measure of the (lack of) slack in the economy, and it may not be unreasonable to think of it as the negative of the unemployment rate, for example. The classic version of the Phillips curve isπt=κxt+εt,leaving away that expectation term. There is a bit of a sleight of hand in much in that literature, in terms of deciding which term belongs on the left-hand side of the equation and which on the right-hand side. Econometricians may demand that the error term should be orthogonal to the right-hand side variable, and that may be the most benign interpretation for the reasons of writing the equations as stated. In section III.A the paper never introduces such error terms, though, which renders that reasoning mute and, in my view, a bit more honest. For example, one can rewrite the classic version of the Phillips curve equivalently as:xt=1κπt−1κεt.Now, the Phillips curve becomes the policy menu that spooks around in so many policy debates: create a bit more inflation on the right to get a bit more xt and thus, say, a little less unemployment on the left. Conversely, the version of the equation stated previously create the monetary-fiscal interaction: a bit more xt, say through fiscal spending, may trigger higher inflation and, in turn, through the Taylor-rule-type reaction function of the central bank, result in higher interest rates. At the zero lower bound, the central bank interest rate is too high to begin with, though, so higher inflation leaves nominal interest rates unchanged, lowering real rates instead. These lower real rates then in turn help to improve the economy, as expressed by some version of the IS equation. Put differently, fiscal stimulus may improve the economy, if one is able to ride the Phillips curve to higher inflation, lower real interest rates, and lower unemployment.If I did not succeed in stating all this sufficiently well, I shall leave it to others to explain this reasoning in a more sensible manner (if it can be done). Here, instead, I wish to ask: Is there a Phillips curve at all? Put differently, is there a clearly discernible relationship between, say, unemployment and inflation or would the search for such a relationship mainly have to point to the noise component and εt? If it is the latter, then the key mechanisms at the heart of the New Keynesian approach have little to do with the forces driving inflation in the data. I submit that this should lead any sensible person to question whether we are barking up the right tree here.I fear things do not look well for the New Keynesian approach. The classic Phillips curve relationship is shown in figure 1. Let me give this relationship a more appropriate name and call it the “Phillips cloud” instead; certainly, that is what it looks to me. Fiscal stimulus ends up poking around in that cloud rather than riding a curve. Sure, there is no correction for the expectation term, as demanded by the New Keynesian version, and one can torture the data in sufficient ways to make it cough up a more visible downward sloping line. Alternatively, one may be encouraged to search for more trustworthy relationships elsewhere. I feel that this will be more productive.Fig. 1. The Phillips cloudSource: The BLS and BEA.Note: GDP deflator, annualized, average percentange change.View Large ImageDownload PowerPointCausal Interpretations of EquationsIndeed, let us talk about the causal interpretations of equations a bit more. The paper here follows the well-trodden path of many papers in that literature, so it is hard to fault the authors. As an example, consider their equationyDt−yDt*=gy(1−ωg−ωg*)(gt−gt*)+ετt+cy(1−ωc−ωc*)(ct−ct*),which the authors then read as stating that “home relative output … depends on three factors—home relative government spending, the terms of trade, and home relative consumption.” One should bear in mind, though, that one could also write this equation as:ct−ct*=1cy(1−ωc−ωc*)(gy(1−ωg−ωg*)(gt−gt*)+ετt−yDt−yDt*).Now, home relative consumption seems to depend on the stuff on the right-hand side. Which causal interpretation is correct? In the absence of stochastic shocks and the ability to then sort the variables in some causal way, I believe that one ought to regard these variables to simply obey a simultaneous system of equation. There is no sense in which the left-hand side of some equation “depends” on the right-hand side. Statements such as the one just cited may be helpful, on occasion, for building intuition, and I applaud the authors for trying to give the readers a helping hand here. It is important to keep in mind, though, that these are ultimately arbitrary interpretations of the equations and that they can be highly misleading.Price and Wage Markup Shocks Are the New Keynesian Explanation for InflationEven if one looks at the data through a perfectly mainstream New Keynesian model, the economic forces on inflation at the center of the fiscal-spending mechanism in this paper and elsewhere in the New Keynesian literature do not really want to show up. In Fratto and Uhlig (2014), we use the standard and well-known Smets and Wouters (2007) model to account for the movements in inflation. This is meant to address the lack-of-deflation puzzle for the postcrisis years, which a number of authors have raised. It turns out that price and wage markup shocks alone account for nearly the entire movement of inflation for the entire sample period and not just postcrisis, and that these shocks account for very little else. In essence, this benchmark New Keynesian model implies that inflation is marching to its own drummer. If anything, the postcrisis developments are slightly better news for the New Keynesian plumbing: there, the zero lower bound and the resulting tightness of monetary policy has at least a small depressing effect on inflation, counteracted by positive markup shocks to produce the slightly positive inflation that we see. One can be happy with that reading of the data, or one can feel rather uncomfortable with our lack of a deeper understanding of inflation and call for a new paradigm, as Hall (2011) has done. Either way, the idea that the relationship between fiscal spending and inflation is really crucial seems far fetched, and the New Keynesian paradigm seems to be of little practical help here. But this relationship and the ensuing impact on monetary policy, depending on whether it is at the zero lower bound or not, is at much of the heart of the New Keynesian fiscal stimulus debate.New Keynesian Models Are DiscontinuousThis paper, along with much of the New Keynesian literature on fiscal stimulus at the zero lower bound, essentially examines the following model stated in Cochrane (2015) and building on the beautiful analysis by Werning (2011). Werning’s (2011) and then Cochrane’s (2015) model is a continuous time perfect foresight simplified New Keynesian model, in which a liquidity trap lasts until some period T in the future. Afterward, the Taylor rule works again and instantaneously: assuming away shocks and output gap dynamics, πt≡0 for t > T. During the liquidity trap, the dynamics are described byx˙t=−σ−1(rt+πt)π˙t=ρπt−κ(xt+gt),where xt is the output gap, πt is inflation, and gt is government spending. The beauty of the approach by Werning (2011) is that one can now characterize the dynamics exactly. Cochrane (2015) then shows what happens as stickiness of prices is driven to zero (see figure 2 taken from his paper). As stickiness disappears, the reaction of the economy explodes: in particular, the fiscal multiplier diverges to infinity. Now, clearly something is wrong here. It surely cannot be right that one can enjoy fiscal multipliers way above 100 for a year or two, while the economy is stuck at the zero lower bound, but prices are nearly completely flexible. Should one then doubt the fiscal multipliers that emerge from this calculus, when price stickiness is calibrated to more conventional values? One may wish to argue that the model shown above or the model analyzed in Blanchard, Erceg, and Lindé is a local linear approximation at that point, and that there is no reason to expect this local linear approximation to work well when prices are nearly not sticky at all. As researchers analyzing a model at hand, that may be a perfectly fine defense, but as a guide to policymakers, robustness checks and understanding parameter variations are paramount. For that, it is important to understand what the model has to say, as the parameters gradually approach the frictionless limit. Let’s just all agree that figure 2 may be the right answer mathematically, but cannot be the right answer for a policymaker interested in practical advice. So then, what is the right answer for the latter? If we do not have a good answer, shouldn’t we then feel a bit uncomfortable with the recommendations emerging from the parameter configuration at hand? I, for one, believe we should indeed.Fig. 2. Cochrane’s discontinuity in New Keynesian modelsView Large ImageDownload PowerPointReal Interest Rates, Policy Uncertainty, and Price StickinessLet me dig a bit deeper into the economics of the New Keynesian profiscal stimulus arguments, as in the paper at hand. The starting point is that real interest rates are too high due to zero lower bound considerations. Fiscal stimulus will then lower them by encouraging higher inflation. Lower real interest rates mean that households and firms get to borrow at lower real rates, stimulating the economy.So let us pause for a moment: Does this argument sound right? It is my impression that safe real rates are really low right now in the Eurozone. Inflation is still somewhat positive, while long-term nominal interest rates on German bunds have fallen to near zero, and some short-term nominal rates are even negative. Real interest rates are negative in the Eurozone. Seriously, shouldn’t safe borrowers be in heaven? Conversely, if real rates that low are not doing the trick, how much lower do they need to be? And why would anyone put an iota of confidence in the argument that additional fiscal stimulus, through its effect of inflation, can get them there? My impression, rather, is that many firms do not borrow due to the uncertainty of the policy environment, or that they can borrow only at high rates due to various risk premia charged by their banks. Or, perhaps, some version of secular stagnation makes firms not want to borrow at all.With lots of optimism and belief in New Keynesian plumbing, fiscal stimulus may move that base rate via that cloudy inflation-output-gap trade-off discussed above. But, surely, this is second order to the bigger problems that these firms are facing. But then why do so much heavy lifting and investment of political capital to focus on fiscal stimulus? It may be better to focus policy attention elsewhere, instead. It may help, for example, to reduce policy uncertainty. The Brexit instead created political uncertainty. Just that one decision appears to have a much larger impact on the Eurozone economy than any German fiscal stimulus possibly could. On the contrary: German fiscal stimulus debates may even create more, rather than less, harmful uncertainty. Let’s start worrying about what matters!The formulation of price stickiness is crucial. The most widely used assumption in the New Keynesian model is Calvo stickiness, where firms are allowed to change their price with some probability in any given period, or have to keep it constant (or indexed) otherwise. This results in a beautifully simple and elegant analysis. However, there is lots of evidence that price changes are state dependent instead (see, e.g., the research by Nakamura et al. [2016] or Alvarez, Lippi, and Passadore [2016], in this volume), as well as the New Keynesian skepticism by Nekarda and Ramey [2013]). Given their fine research, we ought to rely on these approaches to understand what happens in deep depressions or in the face of large fiscal interventions. With deep depressions, for example, it is a bit far fetched to believe that firms stubbornly stick to prices that they chose a long time ago, when everything was rosy. Whatever are their menu costs for changing prices, they will surely be swamped by the advantages of doing so in these situations. Now, granted, the situation in Greece, in Spain, and elsewhere is dire. We need to understand it, and we need to develop sensible policy options. I just find it hard to fathom that stubbornness to stick to prices chosen in good times long gone really is the root problem, and policy interventions that seek to remedy that particular problem may end up chasing ghosts and risk doing more harm than good.VARs: The Inflation ResponseA short remark on the empirical VAR evidence is in order. Figure 3 shows the impulse response of prices to a monetary tightening, as well as fiscal stimulus. The response to a monetary tightening exhibits a substantial price puzzle. As a consequence, I am doubtful that theirs really is a monetary policy shock. I feel there are better ways to proceed: sign restrictions as in Uhlig (2005), say, or various high-frequency approaches currently popular in the literature. I applaud the authors, though, for connecting their theory up to the time series evidence. This is not routinely done in the literature, but I feel it often should be.Fig. 3. The price impulse response to a monetary tightening and a government spending shockNote: Note the increase in the price level following a presumed surprise monetary tightening, that is, note the price puzzle arizing from the identification used by the authors.View Large ImageDownload PowerPointEvidence on Fiscal MultipliersThis paper proposes fiscal expansion to address the challenges in Europe. How successful might that be? In any case, what is the empirical evidence? Quite a bit of current research has gone into addressing these issues, with mixed and murky results (see, e.g., Blanchard and Perotti [2002]; Mountford and Uhlig [2009]; Barro and Redlick [2011]; Ramey [2011]; Auerbach and Gorodnichenko [2012]; Namakumara and Steinsson [2014]; Ramey and Zubairy [2015]; and the discussion and survey by Whalen and Reichling [2015]). It may not be surprising that it is hard to settle these issues decisively, using time series analysis alone. For my taste, I find the analysis of Ramey (2011) and Ramey and Zubairy (2015) and therefore their results to be the most convincing. In a nutshell, they find positive but rather modest fiscal multipliers, even during episodes characterized by a version of a liquidity trap. These empirical multipliers ought to be compared to the core multipliers, calculated in the paper at hand. The zero lower bound moves these core multipliers, with the periphery multiplier staying at a practically constant distance below them (see figure 4). So, if the core multipliers are low, as the empirical evidence suggests, the periphery multipliers are even lower. These results present a further challenge in convincing a skeptic that fiscal expansion in the core of the Eurozone is going to be particularly wise and helpful.Fig. 4. Core and periphery fiscal multipliers without or with a liquidity trapView Large ImageDownload PowerPointOn the Approach and InterpretationReading this paper, one may wonder why the authors did not simply follow some social planner approach instead: formulate the objective, maximize that objective subject to feasibility and incentive constraints, and discuss how to implement the resulting mechanism. Or, more restricted, constrain the planer to use certain tax instruments. But we know from Correia et al. (2013), that such Ramsey problems easily lead to fiscal devaluations as solutions, in order to solve the relative price problems arising from regionally sticky prices. One can keep on constraining the problem until the desired policy conclusion emerges, of course. But now it has become unclear what is assumption, and what is result, and what rules one ought to follow, generally. This quickly becomes a debate about various third-best approaches. What is now a viable argument in the discussion, and what is not? I should not criticize this too much, though. This paper is far better, of course, than the beat-them-dead argument one sometimes hears from various “practical” economic advisors or economic commentators, that some approach proposed by academic economists is not politically viable, while theirs is, of course. Well, how should one know? At least, the paper writes down a coherent model and analysis, and one can then go and constructively agree or disagree on that basis.And one can disagree on the political viability, too, of course. Germany may be the last country with some Eurozone enthusiasm, but recent opinion polls and elections show this enthusiasm to be waning. It will not be easy to convince the German voters that Germany ought to participate in costly monetary or fiscal policy measures aimed first at relieving the dire situation in Greece and elsewhere. Often, Chancellor Merkel has been criticized by the foreign press or by economists such as Krugman and others of not having assumed a bolder leadership position in Europe, of not jumping at the chance of fiscal union, after bailing out the southern states. But the truth of the matter is that Merkel and her party need to be careful themselves to not lose the support of their increasingly disillusioned German voters. Put differently, nobody really expected Great Britain to do much about the situation in Southern Europe, except for dumping the issues on the rest of Europe, per exiting from the mess. It would be good to heed that warning. While Germany is not Britain, it would be naive to count on Germany to be willing to be the residual guarantor for the rest of the Eurozone, for whatever mess arises elsewhere in the future.ConclusionsI would not be surprised if quite a number of readers view the punchline of the Blanchard, Erceg, and Lindé paper as follows: Let Germany fiscally expand at home. This will help Greece.As a piece of research, I admire it: it is excellent. It is ambitious, lucid, informative, and mainstream. My problem is that this New Keynesian mainstream appears to flow into the Dead Sea. Is there really much of a reliable Phillips curve? Are the inflation and real rate responses to fiscal stimulus plausible? Is the safe real rate truly too high right now? Do retailers have a problem because they are stuck with prices from better times? Is that really the first-order problem that needs solving? What about the fiscal multiplier discontinuities in these models, or the less-than-encouraging empirical evidence regarding fiscal multipliers?I find myself in an odd place. These types of models are beautiful. And it is a good rule that only a model beats a model! I wish I had a good alternative at hand. I do not. So my critique is cheap. Mea culpa! Nonetheless, I fear that there are too many holes in this paradigm to trust it for providing robust and sensible policy advice. I agree that these authors are putting one possible version on the table, and that it is worth discussing. Nonetheless, I feel that a good degree of skepticism needs to remain. The proposal to let Germany do lots of fiscal spending at home, with the purpose of somehow thereby improving matters in Greece, would be a hard sale in Germany, in any case.In sum: as a piece of informative and well-crafted economic analysis, thinking through one particular policy option and its mechanisms, it is truly excellent. But if you read it as providing strong support for a policy toward core fiscal stimulus in Europe, then reading my remarks may induce you to become considerably more skeptical.EndnoteI have an ongoing consulting relationship with a Federal Reserve Bank, the Bundesbank, and the ECB. For acknowledgments, sources of research support, and disclosure of the author’s material financial relationships, if any, please see http://www.nber.org/chapters/c13785.ack.ReferencesAlvarez, Fernando, Francesco Lippi, and Juan Passadore. 2016. “Are State and Time Dependent Models Really Different?”In NBER Macroeconomics Annual 2016, ed. M. Eichenbaum and J. Parker. Chicago: University of Chicago Press.First citation in articleGoogle ScholarAuerbach, Alan J., and Yuriy Gorodnichenko. 2012. “Measuring the Output Responses to Fiscal Policy.” American Economic Journal: Economic Policy 4 (2): 1–27.First citation in articleCrossrefGoogle ScholarBarro, Robert J., and Charles J. Redlick. 2011. “Macroeconomic Effects from Government Purchases and Taxes.” Quarterly Journal of Economics 126 (1): 51–102.First citation in articleCrossrefGoogle ScholarBlanchard, Olivier, and Roberto Perotti. 2002. “An Empirical Characterization of the Dynamic Effects of Changes in Government Spending and Taxes on Output.” Quarterly Journal of Economics 117 (4): 1329–68.First citation in articleCrossrefGoogle ScholarCochrane, John H. 2015. “The New-Keynesian Liquidity Trap,” Working Paper, Hoover Institution. http://faculty.chicagobooth.edu/john.cochrane/research/ papers/zero_bound_2.pdf.First citation in articleGoogle ScholarCorreia, Isabel, Emmanuel Farhi, Juan Pablo Nicolini, and Pedro Teles. 2013. “Unconventional Fiscal Policy at the Zero Bound.” American Economic Review 103 (4): 1172–211.First citation in articleCrossrefGoogle ScholarFratto, Chiara, and Harald Uhlig. 2014. “Accounting for Post-Crisis Inflation and Employment: A Retro Analysis.” NBER Working Paper no. 20707, Cambridge, MA.First citation in articleGoogle ScholarHall, Robert. 2011 “The Long Slump.” American Economic Review 101 (2): 431–69.First citation in articleCrossrefGoogle ScholarMountford, Andrew, and Harald Uhlig. 2009. “What are the Effects of Fiscal Policy Shocks?” Journal of Applied Econometrics 24 (April): 960–92.First citation in articleCrossrefGoogle ScholarNamakumara, Emi, and Jón Steinsson. 2014. “Fiscal Stimulus in a Monetary Union: Evidence from US Regions.” American Economic Review 104 (3): 753–92.First citation in articleCrossrefGoogle ScholarNamakumara, Emi, Jón Steinsson, Patrick Sun, and Daniel Villar. 2016. “The Elusive Costs of Inflation: Price Dispersion During the U.S. Great Inflation.” Working paper, Columbia University.First citation in articleGoogle ScholarNekarda, Christopher J., and Valerie A. Ramey. 2013. “The Cyclical Behavior of the Price-Cost Markup.” NBER Working Paper no. 19099, Cambridge, MA.First citation in articleGoogle ScholarRamey, Valerie A. 2011. “Can Government Purchases Stimulate the Economy?” Journal of Economic Literature 49 (3): 673–85.First citation in articleCrossrefGoogle ScholarRamey, Valerie A., and Sarah Zubairy. 2015. “Are Government Spending Multipliers State Dependent? Evidence from Canadian Historical Data.” Working paper, University of California, San Diego.First citation in articleGoogle ScholarSmets, Frank, and Raf Wouters. 2007. “Shocks and Frictions in US Business Cycles: A Bayesian DSGE Approach.” American Economic Review 97 (3): 586–606.First citation in articleCrossrefGoogle ScholarUhlig, Harald. 2005. “What are the Effects of Monetary Policy on Output? Results from an Agnostic Identification Procedure.” Journal of Monetary Economics 52:381–419.First citation in articleCrossrefGoogle ScholarWerning, Iván. 2011. “Managing a Liquidity Trap: Monetary and Fiscal Policy.” NBER Working Paper no. 17344, Cambridge, MA.First citation in articleGoogle ScholarWhalen, Charles J., and Felix Reichling. 2015. “The Fiscal Multiplier and Economic Policy Analysis in the United States.” Working Paper no. 2015-02, Macroeconomic Analysis Division, Congressional Budget Office.First citation in articleGoogle Scholar Previous articleNext article DetailsFiguresReferencesCited by NBER Macroeconomics Annual Volume 312016 Sponsored by the National Bureau of Economic Research (NBER) Article DOIhttps://doi.org/10.1086/690248 © 2017 by the National Bureau of Economic Research. All rights reserved.PDF download Crossref reports no articles citing this article.