Who spent the 2001 Bush tax rebate?

Do tax rebate such as those implemented at various times by the Bush II administration work? Measuring this is not obvious. Previous studies have typically exploited the timing of the receipt of the rebate checks to see how expenses have changed. But most people have anticipated these payments, thus the marginal propensity to consume is mismeasured: it measures the propensity to consume due to short-term liquidity considerations beyond the consumption response from the announcement of the program.

Greg Kaplan and Giovanni Violante go a step further in this analysis and build a model that replicates the measurements found in the literature and the large share of hand-to-mouth households using an economy of liquid and illiquid assets with transaction costs. They define hand-to-mouth households as those who hold less that half their periodic pay in liquid assets. That seems very shaky to measure, as the timing of the relevant survey matters a lot here. But assuming there is no systematic error, they then extrapolate through the model what the response from the announcement of the rebate should have been. This adds 7-8% to the marginal propensity to consume. Interestingly, this come in large part form rich household who have only little liquid wealth because their assets are mostly in real estate and retirements funds. An important consequence of this is that larger tax rebates would have little impact, as they would make it more interesting to bear the costs of putting them into illiquid wealth. In fact, the marginal propensity to consume could even turn negative.

The impact of fiscal uncertainty

Current US fiscal policy is absolutely frustrating. There does seem to be a clear direction, in particular because policy making is rather irrational due to a set of unduly influential and crazy lawmakers. In the end, this means considerable uncertainty about future fiscal policy, in particular because it may not react to economic events in ways that make economic or historic sense. What is the impact of such uncertainty?

Jesús Fernández-Villaverde, Pablo Guerrón-Quintana, Keith Kuester and Juan Rubio-Ramirez address this with a New Keynesian business cycle model that feature variable volatility in fiscal policy. Their conclusion is that the current uncertainty lowers activity and has the policy equivalent of a 25 basis point increase in the federal funds rate, which I find rather minor. The model rightfully yields that the main mechanism is through investment and the uncertainty on capital return taxation. I find it interesting that it leads to stagflation, as firm opt for higher prices to reduce miss-pricing costs. In the end, the authors show that if one removes the usual automatic stabilizers and assume very persistent fiscal shocks, which may be a good characterization of the current situation, the prediction is a 0.5% reduction in output, which I am ready to believe.

Public consumption and the business cycle

One aspect of government purchases the current crisis has highlighted is how volatile they can be. Quite obviously, they are influenced by politics, to the point of complete reversal between massive spending and severe belt-tightening within months as in the US and the UK. But there could also be a more systematic component that is linked to the business cycle. After all, the government may be trying to improve the welfare of its constituents and for example substitute public consumption for lacking private consumption, or the same for investment.



Ruediger Bachmann and Jinhui Bai look at this using an augmented real business cycle model. They claim that 25-40% of the variance of public consumption can be accounted for by shocks to total factor productivity once implementation lags and costs of public consumption, as well as taste shocks to public vs. private consumption. I am no particular fan of taste shocks, as they are the symptoms of a modeler who is giving up on trying to explain something and simply equates the error term in the Euler equation to a shock. Then much is driven by how this shock is calibrated, in this case to match a four year electoral cycle and some data moments. When I think about shocks in this context, I think indeed about who is in power to decide on public expenditures. But that is not completely exogenous. Indeed, the state of the economy has an impact on who gets elected or reelected. And this can be calibrated without trying to match the data moments one is trying to explain.

Convergence in recessions

Growth theory and data teach us that, at least in developed countries, economies tend to converge in the long run: the dispersion across regions or nations of per capita income (or similar indicators) tends to decline. While this is a long term phenomenon, there is a priori no reason to believe this is a constant process.



Eldon Ball, Carlos San Juan and Camilo Ulloa study total factor productivity in agriculture across US states. While they indeed find a general trend towards convergence, it turns out that its speed is much faster during recessions. Why would this happen? If we follow Schumpeter, the worst firms should be dropping out during a recession, thereby relatively increasing TFP in the worst areas. And voilà, you have faster convergence. But only farm-level data would tell whether my conjecture is true.

Policy risk and the business cycle

The US economy seems stuck in its tracks, and many blame uncertainty about future public policy, including me. Indeed, private firms are currently sitting on a lot of cash and are making very good profits, yet they are not investing or hiring. This really looks like a wait-and-see game. But it this justification well-founded or is it just a cheap excuse to justify higher than usual profits in the face of high unemployment?

Benjamin Born and Johannes Pfeifer put some structure into these arguments by taking a standard New Keynesian model and adding uncertainty about monetary and fiscal policy. They measure this by looking at tax rates and monetary policy shocks with time-varying volatility. Previous literature already looked at the impact of aggregate uncertainty, which policy makers can do little about. But policy uncertainty is another matter. And there is hope, as Born and Pfeifer show that the impact of policy uncertainty is not that important (but much larger than uncertainty about productivity shocks) thanks to monetary policy reaction through a Taylor Rule. So that is somewhat reassuring, but then the size of the current policy uncertainty is an order of magnitude larger than when this paper was written, and monetary policy is bound by non-negative nominal interest rates.

Fiscal policy as insurance

The goal of fiscal policy is at the macroeconomic level to steer the economy towards efficiency and, depending on the country, to smooth somewhat economic fluctuations. It has long been debated whether this is desirable or possible at all, given the large delays in implementing public expenses. But changes to tax policies are quicker to put in place and implement. At the microeconomic level, the focus is more on the long term, again try to attain better efficiency as well to optimize some definition of fairness across economic agents, however this may be defined in the respective countries. These micro and macro aspects have largely been regarded as separate. This does need to be so.

Eduardo Engel, Christopher Neilson and Rodrigo Valdés look at the particular fiscal policy of Chile. This country is characterized, like many emerging economies, by wild fluctuations in economic activity. In this case this is triggered by changes in commodity prices, in particular for copper. The most important implication is that government revenue varies wildly (a macroeconomic impact) between 1 and 8% of GDP, which changes Chile's ability to redistributes across heterogeneous households (a microeconomic impact). Adhering to a balanced budget rule would have a dramatic effect, in terms of aggregate welfare it would be like renouncing to half of the copper revenue. The reason is that households' incomes is also correlated with copper revenue, and a countercyclical policy is then optimal. And to be the most effective, the poorest households are helped in hard times, both because they have the highest marginal utility from consumption and because they have the highest propensity to consume.

Chile has been pursuing so far something that is close to a balanced budget rule: expenses are related to a permanent income measure of income. This means expenses are relatively constant, except for the last years, where expenses grew significantly despite a reduction in copper prices. This appears to have worked well, in particular because the poor have been the target of this largesse, not the rich. That was stimulus spending done right. This paper shows how this can be done even better.

What is a sticky price?

An amazing amount of scholarly effort is devoted to figuring out optimal stabilization policies in developed economies. I am not convinced this effort is well-placed, as fluctuations in developing economies are much larger and long-term trends quickly swamp short-term fluctuations in welfare assessment for developed economies. The last recession in the US may make it worth to look at stabilization though.

Greg Mankiw and Matthew Weinzierl have a piece of rather pedagogical nature trying to convince us that stabilization policy is worthwhile. Their model is essentially the one that is taught to undergraduates: a two-period model with households maximizing intertemporal utility from consumption, a government, and firms that maximize discounted profits. Oddly, firms do not care about the resale value of capital in the second period, which makes investment largely irrelevant. Finally, prices are fixed the first period, but can be changed in the second period.

Beyond the pedagogical merit, can this model be used for serious policy prescriptions, which Mankiw and Weinzierl even quantify? For one, the last recession was sufficiently important that prices and wages actually adjusted down in the short term, which violates the critical premise of the model. Indeed, all what policy tries to do is undo the frictions stemming from price rigidity. Second, when prices do indeed not change in the short-term, it is presumably when it is not worth doing do so, thus policy intervention also does not seem worth it. Of course, it could be that there is a genuine Keynesian lack of demand, but this can be attacked best by dealing with what causes the lack of demand, not by creating artificial demand through government expenses. For the last recession, this would have been easing collateral constraints. Third, the model assumes a money quantity equation, which imposes a constant money velocity. I thought we all had agreed long ago this was a silly assumption.

I really do not understand the point of this paper. After all, as Mankiw likes to say on his blog, all this can already be found in his favorite textbook.

Socialist economies smooth better the cycle

Capitalism is often presented as a wild economic system where conditions are harsh as everyone fights for his survival. The fact that economic agents are not sheltered against shocks leads them to be more efficient and possibly protect themselves better against events. Incentives are not as well aligned in a socialist economy, as free-riding is more prevalent and weaker agents may be more likely to survive in such a sheltered system. The endless discussions on which system is better ultimately boil down to preferences about risk tolerance and fairness, and on which system offers higher welfare.

Bruno Amable and Karim Azizi point out that more socialist economies appear to be better at smoothing out business cycles in the aggregate. Indeed, they tend to adopt more readily Keynesian policies, which do smooth somewhat economic fluctuations, France being the prime example. But that does not yet mean these economies are better: while fluctuations are lesser, the average level may also be lower. And fluctuations may be optimal, as we have learned from the real business cycle literature. So the jury is still out.

Crime on the job and the business cycle

The cyclical behavior of work effort is rather puzzling. One would expect that people would work harder during a recession to avoid getting laid off, yet measures of labor productivity (per worker or per hour) are consistently positively correlated with GDP. This also runs counter to the argument that the least productive workers are laid off first in a recession, which should improve the productivity of the remaining ones through a composition effect. Survey data is much more mixed, though, but that is often based on perceptions rather than facts.

One reason why labor productivity may vary could also come from counterproductive efforts from the workforce: stealing, sabotaging, annoying co-workers. Aniruddha Bagchi and Siddhartha Bandyopadhyay fold all these activities under the crime label and ask whether this is linked to the business cycle. There is no data about this, unless you think like the authors that this is only dimension that makes labor productivity vary. So you are left with purely theoretical exercises. The authors highlight here to contradictory effects. First, pretty much everyone gets a job in a boom, including those "criminals," which would lead to a negative correlation of labor productivity with output. But this effect could go the other way if labor market prospects are likely to weaken and jeopardize re-employment. Second, they assume that deviancy requires a setup cost, which one is less likely to bear when the labor market weakens. This would even reinforce this negative correlation.

This possible ambiguity would need to be sorted out with a tight calibration exercise at least, or some structural estimation with hidden variables. But the authors just wave hands and claim things can go either way. In any case, they are probably right not to pursue. Using a two-period model to study business cycles is silly anyway.

Search effort under mass unemployment

As discussed previously here, one good reason for prolonging the duration of unemployment insurance insurance coverage in a deep recession like the last one is that it would be unfair to expect from the unemployed workers to find as easily a job as in normal circumstances. Or, in other words, even if they apply the same work effort, they have a smaller chance of finding a job given the labor market tension and should be allowed to be protected for a longer time.

Alan Krueger and Andreas Mueller study what happens to search effort when there is a smaller change of finding a job. Specifically, they interviewed several thousand unemployed weekly during the last recession about their reservation wage and their search effort. They find that the reservation wage is essentially constant, expect for older workers with sufficient cash reserves, but workers are willing to go below that reservation wage for part-time work. The time devoted to search, however, dips quickly over the unemployment spell. That should not be surprising given how little time people spend looking for a job (from a study by the same authors). What is more interesting is how all this compares to a normal recession. Unfortunately, Krueger and Mueller offer no discussion in this regard.

The impact of the extension of unemployment insurance benefits in the US

Given the depth of the last recession and the obvious difficulties unemployed workers have to find new jobs, the US government has successively and temporarily extended the usual 26 week period during which unemployment insurance benefits are given, up to 99 weeks for some workers. On consequence that has been worrying some is that this will leads job seekers to seek less jobs, as there is less urgency to be employed. But pointing to the longer duration of unemployment is not appropriate, as this depends to a (very large?) extent on the general business climate. Just looking at data is not sufficient, you need to put in some structure in the form of a theory.

Makoto Nakajima builds an elaborate model that features job search à la Mortensen-Pissarides with variable search effort, consumption-saving decisions, borrowing constraints, skill depreciation in unemployment and appreciation in employment. This complex model is then calibrated to the average state of the US economy, and set to start in a state as close as possible to the one in 2007. Then the transition paths are computed as the economy is hit by shocks, with and without benefit extensions, assuming economic agents did not expect the extensions. All in all an impressive exercise. The conclusion: about a quarter of the 4.8 point increase in the unemployment rate is due to the longer duration of benefits. This is not insignificant, but it could be an acceptable price to pay for the exceptional circumstances.