Dealing with the Evidence

The recent challenge to a 2010 paper by Carmen M. Reinhart and Kenneth S. Rogoff, authors of the widely acclaimed This Time is Different, has seemingly dealt a fatal blow to the case for fiscal austerity. The most well known feature of Reinhart’s and Rogoff’s research is their “debt threshold” figure, which finds that at 90 percent debt-to-GDP the average country begins to experience a negative growth rate. This has been shown to be wrong, although there is dispute on the authors’ data manipulation. Today, most major developed nations, burdened by recession, are either approaching or have surpassed a 90 percent ratio, as such Reinhart’s and Rogoff’s empirical approximation of the “threshold” seems to carry some importance. In reality, their debt cut off point probably has very little importance, and its empirical refutation does not hurt the more nuanced case for austerity. Likewise, other recent empirical research on austerity and the fiscal multiplier — the value created by each additional dollar spent by government — should be interpreted carefully, because it is not obvious that the surveyed data represents actual cases of austerity.

For what it is worth, the claim that Reinhart and Rogoff are representatives of the case for austerity is highly questionable. Their detractors accuse them of getting the causation wrong, suggesting that it is slow growth which leads to high debt-to-GDP ratios, as opposed to vice versa. However, in most of their academic work, they are careful to point this out to their readers. No less, their principle argument, that debt is not costless (see Reinhart, Reinhart, and Rogoff [2012]) — even when treasuries yield a nearly zero percent return —, has mostly been ignored. It is quite ironic, because they have been accused of being sloppy in their research. This claim, as we have seen, is true, but much of their work can be interpreted as a means of pushing for greater nuance in the debate on debt.

The debate on public debt is not the only area where there is a lack of quality. Debt is just one facet of the problem, and, arguably, not the most important. All economic recoveries require that the private sector recover from the shock of resource misallocation, and for this to occur the institutions that govern the market must avoid setting obstacles that delay a beneficial recovery in private investment. In reality, public policy has failed to fulfill this role. It is not surprising that countries where “austerity” has performed particularly poorly are the same as those which suffer from markets with overbearing government-imposed transaction costs.

The case for austerity comes with a specific vision of the market process that often gets lost in translation. Most models are relatively general, because they necessarily abstract from most features of the real world. Economists build these models to understand the causal effects of certain actions and interventions. They are theoretical alternatives to controlled experiments, which are extremely difficult (even impossible) to conduct when studying complex phenomena. But, when it comes to policy recommendations, these models tend to be unsuitable for the task. The real world does not respect the abstraction of the model, but is instead an outcome of many causal processes occurring simultaneously. There is no ceteris paribus, so, while theory is absolutely indispensable for understanding actual economies, you need to consider the totality of theories to get an accurate picture of a live market process.

Markets are composed of individuals, each with certain ends in mind and a specific set of means to achieve these ends. Each action has costs and benefits, and the individual will have to choose which route to take based on these. As established by Mises (Mises [1920], Mises [1922]), these costs and benefits become known to us through price formation. It follows that changes to any change to these prices will also change the decisions made by individuals. The case against government spending follows from this fundamental insight, because the institution of the market provides the best disciplining process — profit and loss —, which public expenditure is not subjected to. But, public expenditure is only one facet of what “austerity” really means, because the other half of the theory calls for a private sector-led economic recovery. For this to occur, the institutions which govern the market must allow individuals to choose undisturbed and undistorted.

Consider, for example, the case of Spain. While the belief that Spain has been practicing fiscal responsibility is suspect in and of itself, there are bigger problems than public spending. The Iberian country has been suffering from structural problems for some time, even before the recent financial crisis. Spanish labor markets are inflexible, because of the various regulations of employment contracts and employee rights, and this is manifested in the country’s large informal sector. Economist Friedrich Schneider estimates that prior to 2007 Spain’s informal sector was valued at over 20 percent of the country’s GDP. This has made the economy much less competitive. Its growth during the late 1990s and throughout the 2000s was driven primarily by the European credit boom, but much of this growth occurred in sectors that later collapsed. These structural problems were recognized by most economists before the recession, and the expectation was that Spain (and countries in similar situations, including Greece and Italy) would deal with them during the boom. They did not, but after the recession many economists, for whatever reason, placed less weight on them than they did before. But, if Spanish markets are hampered to the extent that only a credit boom can bring alleged prosperity, how can we expect the private sector to recover from a major macroeconomic shock?

The situation is similar in the United States. There are uncertainties regarding the costs of hiring labor, including some past ambiguity on the costs of health care. Changes in the minimum wage will also impact the rate of hiring; an increase in the minimum wage will force firms on the margin to rethink their hiring policies, and may even force them to shed unaffordable employees. Bad monetary policy has also added to the aura ambiguity. First, there is little consideration of how bad monetary policy impacts prices, including making price adjustment more difficult. Second, as economist Jeffrey R. Hummel explains in his contribution to Boom and Bust Banking, the Federal Reserve responded to the crisis by allocating credit. This has impacted the banking system, in part protecting otherwise moribund lenders. Banks, as financial intermediaries, are indispensable to a working economy, as, ironically, argued by Ben Bernanke (Bernanke [1983]). Credit allocation may have saved some financial firms from bankruptcy, but this was done without properly considering the prospective costs — the misallocation of resources towards the financial sector and the creation of ambiguity regarding this sector’s state of health. Finally, much of public policy designed to help induce recovery has targeted consumer spending, whereas the problem is principally one of inadequate real private investment.

The theoretical point is that if economies are an aggregate of several simultaneously operating, interrelated, complex phenomena, an intervention can impede one process and therefore slow down the entire recovery. If interventions of this kind affect a significant number of different processes, then the results can be particularly damaging. This was the case, for instance, with Franklin D. Roosevelt’s New Deal policies, as chronicled by Milton Friedman and Anna Schwartz in their seminal A Monetary History (1963). These kinds of “stimulants” are somewhat analogous to feeding a bodybuilder steroids, but simultaneously tying his arms behind his back, so that he is absolutely incapable of building muscle mass on his own.

The fundamental task to be accomplished during a recovery is the readjustment of the structure of production. This is true whether you ascribe to Austrian business cycle theory, a “demand-side” explanation, or real business cycle theory. It was intermediate supplies, or factors of production, that were most volatile during and after the boom, not consumers’ goods. It was investment which took the greatest hit, if we compare the percent decline of different aggregates. Any recovery must be a recovery in real private investment. Any policy which distorts spending in favor of consumption, at the expense of investment, will make recovery more difficult. Any policy that increases the costs of the employment of capital, including raising the costs of complimentary labor, will make the recovery more difficult. If the pricing process isn’t allowed to adjust because of an extravagant, directed monetary policy, the recovery will suffer.

Many economists will respond to the arguments made so far by arguing that the relative lack of consumption is being ignored. However, apart from the fact that the hit to consumption was relatively minor, it is important remember that present consumption does not decide the scope of present investment. All production is meant for future consumption. The prices of producers’ goods will reflect the expectations of future consumption. We know that the prices of factors of production are imputed from the final product, through the competitive bidding process of the market. If expected future consumption is relatively low, then the price of the factors of production — in the present — will fall. Price rigidity is not an adequate explanation for why this hasn’t happened, or at least why it hasn’t led to a stronger recovery. It has been five years since the financial crisis, if prices have not adjusted then the fault lies somewhere other than the market (public policy).

In more sophisticated circles, the consumption argument is no longer as widely circulated. Is the explanation to be found in interest rates? Instead, many economists have posited that the problem is that the real interest rate is somewhere below zero, and monetary policy has not helped push market interest rates to a sufficiently low level. Why do markets have trouble setting interest rates on their own? A major shock will create greater uncertainty, inducing people to look for relatively liquid forms of savings — assets they can quickly convert into other necessities, when the need for them arises. Less liquid assets must increase their attached “liquidity premium” to draw savings, otherwise people will prefer more liquid alternatives. This will push interest rates above their “natural” position.

There are a few issues with the negative interest rate theory. One concern is that, if interest is primarily decided by social time preference, it implies that people are willing to pay to have a fraction of their income temporarily taken from them. Another is that the creation of a subsidized liquid asset, like a government bond, creates an artificial alternative to other assets, increasing the liquidity premium of non-subsidized assets (which are less liquid because the risk of default is higher). If these subsidized assets did not exist, individuals would have to choose amongst alternative forms of savings, and the liquidity premia on these would be correspondingly lower. U.S. treasuries are often shown as evidence that the natural rate of interest is below zero, because people are essentially paying to invest in short-term treasuries. If the rate of interest on public bonds is artificially low, due to their subsidized nature, then these are not good indicators. Further, since inflation erodes the real value of the dollar (and, thus, the real value of interest), any inflation will seem to lower real interest rates further, simply because the market cannot raise the inflation premium. They cannot compete with the subsidized rates of public bonds during a period in time where there is so much uncertainty.

Ultimately, a flight to public debt creates the dual problem of misallocated resources and a crowded out private market. The misallocation occurs out of virtue of public expenditure: the institutions of public governance are not good at disciplining bad investments. The crowding out is not just an issue of real resources being drawn away from the private sector. It is also an issue of making the private sector uncompetitive, especially the all-important market of financial intermediation. If Gresham’s Law says that bad money drives out good money when the price of the former is held artificially high, we could call the current situation the Gresham’s Law of public debt.

The narrative told here finds its foundations in Robert Higgs’ theory of regime uncertainty (Higgs [1997]). The basic idea is that during a recession uncertainty is an added cost, and any policy which adds to it will raise the price of using the market. This will necessarily hamper a private recovery. Regime uncertainty is ridiculed by other economists, often referred to as the “confidence fairy,” but such characterizations would only be true of an extremely uncharitable interpretation. In reality, regime uncertainty is a theory which considers the complexity of the market process in its totality and recognizes that hampering the individual processes of the market can lead to relatively sub-optimal macroeconomic outcomes. Considering the weaknesses of competing theories, it is these theories that are unsophisticated.

More importantly, returning to the central thread of this argument, the empirical arguments against austerity are biased. Even humoring the notion that developing countries have practiced policies of austerity, which is an already contentious claim, the fact remains that without serious market reform the recovery will remain agonizing. It is no surprise that it is the least flexible markets — e.g. Spain, Greece, et cetera — that are preforming the worst. Studies which track the performance and spending patterns of countries supposedly being subjected to “austerity” (which usually means spending increases under the Keynesian ideal) are not making fair comparisons. What is needed is a lifting of restrictions on the market processes’ allocation of factors of production. So far, this point has been lost to fiscal doves, and we are worse off because of it.

  • http://twitter.com/SFait79 Silvano IHC

    I partially agree. I mean there is a lot of work to do to explain the “European” side of the story.

    First of all, if unflexible markets are relevant, we have to explain why and how they could overborrow and why savings had been allocated there. In particular why countries with an excess of savings lended so much abroad instead of investing at home or in more developing countries. Productivity by itself can be enough to explain growth and real income differentials, but not a financial crisis. If marginal productivity is fairly priced and credit is allocated according to it a country is going to be credit constrained comparded to its potential or just financially underdeveloped. Overborrowing is a different issue. In Eurozone crisis spread out in a very asymmetrical way, even in a country like Ireland which is market friendly while France with its traditional big government performed relatively better.

    In order to explain asymmetries we have to find variables which are (more) present where the crisis hit harder while they were (almost) absent in stronger countries. Better if these variables present a low variance inside each both sample. This is not a simple task.

    Last, but not least, the explanation should be consistent with the historical trend. Even if Spain has its rigidities, the country was on track. I mean, starting from the process of convergence during ’90s many things changed. Indeed Aznar was considered to be one of the most market friendly politicians in Europe. At the same time while Italy privatized state-owned companies for more than 100bn USD, implemented inflation targeting, financial liberalizations and some kind of labour market reforms, French government was much more intrusive than its peers passing even a bill which aimed to reduce to 35 hours per week the working time. Anyway France keep on track with its competitors, Italy failed on that.

    Regime Uncertainty costs are interesting, but if they can’t be assessed even using some sort of proxies, they’re hard to discuss. Every explaination should be enough robust to resist in a satisfactory way cross border comparisons and consistent with a temporal analysis of the trend. Indeed, this is a difficult task.

  • Robert Mróz

    A great post. I was astonished and disappointed to see that so many otherwise savvy people put so much weight on this 90% value, as if any econometric estimation could really yield quantitative and robust results instead of only qualitative ones.

    Also, Eurozone seems to be just buying time again and again, and I’m wondering whether they’re using this time for anything else than trying to figure out how to buy more of it. Structural reforms are indeed the key.

  • Pingback: How Not to Approach Empirical Research | Economic Thought

  • Pingback: NGDP and the Recession | Economic Thought