Stirati on the NAIRU

Surprisingly there was no entry on the Non Accelerating Inflation Rate of Unemployment (NAIRU) in the New Palgrave Dictionary of Economics. There is, and remains there, one on Milton Friedman's Natural Rate of Unemployment, written by Michael J. Pries. Both are deeply interconnected concepts. In fact, Antonella Stirati, who I asked to write it, presents the NAIRU as a reformulation of Friedman's natural rate within models that allow for real wage rigidities and involuntary unemployment.

Despite differences in microeconomic foundations, both concepts imply a vertical long-run Phillips curve and share three central propositions. First, the equilibrium unemployment rate is determined independently of aggregate demand. Also, unemployment below it causes accelerating inflation. Finally, actual unemployment is eventually drawn back toward it.

In the standard wage-setting/price-setting model, the NAIRU is the unemployment rate required to reconcile workers' real-wage claims with firms' desired markup. Labor-market institutions (e.g. unemployment benefits, employment protection and union bargaining power) do not directly determine the real wage, which is constrained by productivity and the markup. Instead, they determine how much unemployment is required to discipline workers sufficiently to make wage claims compatible with the markup.

This framework generates a wage-price spiral whenever unemployment departs from the NAIRU. In New Keynesian versions, stronger aggregate demand first reduces involuntary unemployment. Then lower unemployment raises wage claims, which are assumed to pass fully into prices because the real markup is fixed. Monetary policy subsequently raises interest rates, reduces investment and aggregate demand, and restores unemployment to the NAIRU. Stirati emphasizes that this adjustment mechanism depends on two weak assumptions, one that interest rates reliably control private investment and demand, and then that departures from the NAIRU have a sufficiently strong and predictable effect on inflation.

 

The policy consequences are strongly supply-side. Because the NAIRU is attributed to labor-market institutions, reducing unemployment supposedly requires weaker employment protection, lower unemployment benefits, diminished union bargaining power and, sometimes, greater product-market competition. The whole neoliberal policy agenda. Keynesianism is impossible, since aggregate-demand policy can affect unemployment only temporarily, while permanently changing the inflation rate.

 

Her principal criticism concerns estimation. The NAIRU is unobservable, yet it is central to monetary policy, potential-output calculations and European fiscal rules. In practice, estimates are extracted from the trend of actual unemployment or chosen so that the estimated unemployment gap best explains inflation. Consequently, movements in unemployment that do not produce inflation are simply absorbed into a changing estimated NAIRU and relabeled as "structural." The resulting estimates are highly uncertain, frequently revised and closely track actual unemployment, often without corresponding changes in labor-market institutions.

 

The evidence is really against it. Major historical episodes completely contradict the model predictions. For example, persistently high European unemployment did not produce accelerating deflation. Or the low unemployment of the Clinton boom did not produce accelerating inflation, and the sharp post-2008 rise in unemployment did not cause sustained deflation. This suggests that the estimated NAIRU is less an independent structural attractor than a moving average of unemployment shaped by aggregate demand and capital accumulation.

 

The entry’s central argument is that all three defining properties of the NAIRU, its independence from demand, its role as an inflation barrier and its status as an attractor, are theoretically questionable and empirically unsupported. Nevertheless, the concept survives because it remains embedded in macroeconomic models and policy institutions, especially European fiscal policy.

 

The entry requires access to the Palgrave. But many of the same arguments are discussed in her Godley-Tobin Memorial Lecture, freely available here.

On the irrelevance of inflation expectations: the return of the working class

There are many myths about the Phillips Curve and the so-called Monetarist Counter-Revolution of the 1960s and 1970s. Forder's book is a good read on some of these issues. Jeremy Rudd's recent paper is also a must read, and has now been accepted for publication in the Review of Keynesian Economics (ROKE).

It debunks the myth about the importance of inflationary expectations for explaining the Great Inflation of the 1970s, and casts doubts about the Monetarist Counter Revolution, the Rational Expectations one, and that would include the more recent New Keynesian developments that include a lot from both. Rudd's paper suggests that there is a simpler explanation for the acceleration of inflation back then, and the Great Moderation since the 1990s. In his words:

"Another way of stating this point is that an important feature of inflation dynamics after the mid- 1990s appears to be the lack of a strong wage–price spiral (or of any significant year-to-year feedback between wage growth and inflation)... An observation about the actual nature of the 'wage bargaining process' is helpful at this point."

Old Keynesian stories about cost-push versus demand-pull inflation come back to mind. He suggests essentially that worker's wage bargaining power has been weakened, and that reflects in less impact of labor costs on actual inflation. Again, in his words:

"We can make a reasonable case that reactions of labor costs to actual inflation were an important feature of the inflation process in the 1970s. In particular, without this channel’s being present, it is nearly impossible to explain why movements in food prices left a durable imprint on core inflation—unlike energy, which can plausibly be viewed as a broader input to industry, changes in agricultural prices shouldn’t act like a cost shock to firms outside the food sector."

This suggests a preoccupation of the profession with more concrete, objective causes of inflation dynamics, and that bargaining power of the labor class matters. Martin Sandbu has suggested a few months ago in his Financial Times column that:

"To be precise, we may be witnessing the manifestation of two outmoded ideas: that the relative power of economic classes alters macroeconomic outcomes; and that macroeconomic policy tilts that relative power."

Sandbu also noted that in this recession it is Michal Kalecki, not Keynes, the old and forgotten master that we should bring back from the dead. While I do think that Kalecki should always be remembered and was and continues to be relevant, it is less clear that the wage bargaining position of the working class will recover anytime soon. Sure, if Dems manage to pass the more robust social spending bill, with the extra US$ 3.5 trillion, things would look a bite better. Hope springs eternal.

The Economist and the natural rate of unemployment

The Economist new series on 'big ideas' tackled in a recent issue the concept of the natural rate unemployment (subscription required; other ideas included Say's Law and Human Capital, just to give you the broad picture of what they think it's big). I will only comment very briefly on two issues, one related to the history of ideas and the other to the concept itself.

The Economist suggests that the natural rate of unemployment can somehow be connected to the ideas of Keynes. In their words:

John Maynard Keynes, the great British economist, took a first step towards the natural-rate hypothesis when he focused minds on 'involuntary' unemployment. In his book 'The General Theory'.

The idea of a natural rate of interest can be clearly traced back to marginalist economics, and was a central concept discussed by Knut Wicksell. The idea of the natural rate of unemployment, as The Economist correctly points out, is essentially related to Milton Friedman's American Economic Association Presidential address of almost half a century ago, and the Monetarist interpretation of the Phillips Curve (PC).

However, Keynes' Wicksellian days were over by about 1932, when in the aftermath of the Circus discussions, he abandoned the theoretical framework of the Treatise on Money and started on his way to effective demand and the General Theory (GT). Note that the idea of a natural rate of interest is intrinsically connected to the natural rate of unemployment, since the former would be the rate of interest that corresponds to full employment and stable prices, meaning the equilibrium in the labor market.

On the natural rate of interest, Keynes had this to say in chapter 17 of the GT:

... it was a mistake to speak of the natural rate of interest or to suggest that the above definition would yield a unique value for the rate of interest irrespective of the level of employment. I had not then understood that, in certain conditions, the system could be in equilibrium with less than full employment. 

I am now no longer of the opinion that the concept of a 'natural' rate of interest, which previously seemed to me a most promising idea, has anything very useful or significant to contribute to our analysis.

There is little room for doubt that Keynes rejected the natural rate rather than taking the "first steps" in its direction, as The Economist suggests. He was clearly moving away decisively from that flawed concept, which leads me to discuss the concept itself and what the magazine (they think somehow they're a newspaper, strangely enough) thinks one must do to reject the concept.* The Economist says:

... to reject the natural rate entirely, you would need to believe one of two things. Either central banks cannot influence the rate of unemployment even in the short term, or they can peg unemployment as low as they like—zero, even—without sparking inflation. Neither claim is credible.

The second part shows that they presume that the absence of a natural rate means that the economy could be driven to zero unemployment without inflation (I'll deal with the first statement about the effect of monetary policy on employment subsequently). This is the case because for The Economist, as much as for many in the mainstream of the profession, inflation is always a demand phenomenon. Too much spending, often government (meaning fiscal policy) is behind inflationary pressures. Of course, the PC only works empirically when supply side factors, like the price of oil, are included as explanatory variables. So they think that if you reject the natural rate you must think there is no limit to demand expansion.**

Inflation, more often than not, is a cost push phenomenon associated to higher prices of commodities, devalued exchange rates, and more social conflict reflected in higher wage resistance. So, lower unemployment would eventually strengthen the labor force, and would lead to higher wages, and that combined with possible sectoral shortages of raw materials, which would also be more expensive, would lead to inflation. So you might very well get inflation before you reach zero unemployment rate, but the reasons are not the ones the magazine authors think. The point is that there is no reason to think that there is one level of unemployment that magically would lead to inflation. No natural rate, but several possible rates of unemployment compatible with stable prices, depending on a whole set of social and institutional conditions.

And the central bank can affect unemployment, even if its ability is not straightforward and is asymmetrical. In a recession, a lower rate of interest is, as Eccles famously said (he really popularized the phrase, but he did not coin it), like pushing on a string. You need fiscal spending in that situation, as has been painfully obvious during the long Obama recovery. But in a boom a hike to the rate of interest, in particular if it affects debtors, might cause a recession. So The Economist is wrong (I'm shocked, shocked!). You can believe that central banks affect unemployment and that there is some inflation barrier, but that does NOT mean that there is a natural rate.

Interestingly enough, they do admit that they (and economists in general) have no clue what is the correct natural rate of unemployment, and that should make the natural rate concept of limited usefulness for policymakers. So what they propose central banks should do? Use a flawed concept that cannot be measured as the guide for policy? That's some serious thinking, isn't it?

* Too many posts on the problems with the natural rate have been discussed over the years in this blog, just click on the hashtag and check a few if you're interested.

** Note, also, that when demand goes up, firms do invest to adjust capacity, which means that the capacity limit is not fixed either.

The NAIRU or why economics is not a serious science

This is from Watson's AER paper from last year (another version here). Same methodology he has used before, with Staiger and Stock, as far as I can tell, to measure the non-accelerating inflation rate of unemployment (NAIRU) or natural rate. The estimation is based on the Phillips curve (PC), in which inflation is the result of deviations of unemployment from its natural level, that is, essentially demand. He estimates that it is at 6.3%, but will soon return to its pre-2007 crisis level of 5.5%, basically were we are now.

Couple of things. In the estimation of the PC he does, as normally is done, include supply shock elements. My guess (results are not clearly shown in the tables) is that a good chunk of inflation is actually explained by this rather than the unemployment gap. In other words, cost matters for inflation, and whether unemployment was above 10%, as right after the crisis, or 5.6% like now, has little impact. Also, that means that the unemployment gap is basically orthogonal to inflation, and that the measure of the NAIRU or natural rate could be obtained as an average of the actual unemployment. In one of the methods of obtaining it that is exactly what is done (not in this paper).

In other words, the black line in the middle is NOT an attractor around which the actual unemployment level fluctuates. It is an average of the actual rate, which is essentially another way of writing the same data. It is the actual unemployment rate that determines the natural one, and if unemployment rates were reduced sufficiently with expansionary policies, the natural (being an average) would also come down. Economics is NOT a serious science.

Notes on the Policy Implications of the New Macroeconomic Consensus

The New Macroeconomic Consensus (NMC) model is based on three simple equations. An IS equation that, contrary to what most discussions within the heterodoxy suggest, is based on a Ramsey model intertemporal approach to savings and investment, a Phillips curve (PC) equation, normally with rational expectations, and a monetary policy (MP) rule, typically Taylor’s rule. From the IS and the MP an aggregate demand (AD) curve is derived, while the PC provides an aggregate supply (AS) curve, similar to Lucas’ supply curve. Business cycles are seen as being determined by shocks, either monetary, that affect the AD curve, or real, which impact the AS curve.

A few things are important to note with respect to the NMC model. First, the IS curve now is not based on the traditional Keynesian multiplier process, by which savings adjust to investment (or in more sophisticated models with endogenous investment, to autonomous demand) as a result of variations to the income level. Agents make intertemporal decisions on consumption and savings, and investment adjusts, in the absence of imperfections, to full employment savings as in the pre-Keynesian models. That is the reason why in order to stimulate the economy it is often suggested that what is needed is higher inflationary expectations (which in this framework could be caused by the central bank announcing a higher inflation target), which would in turn lead to an increase in current consumption (since inflation would reduce future consumption possibilities; see my critique of this view, which I refer to as the inflation expectations fairy, here).

Second, both the New Keynesian Phillips Curve and the Taylor rule presuppose the existence of a natural rate of unemployment, in line with Milton Friedman. Further, stabilizing the rate of inflation around its target is tantamount to stabilizing output around its full employment level, a result sometimes referred to as the ‘Divine Coincidence.’ Thus, inflation is always the result of a level of unemployment that is below its natural level, or in other words caused by demand-pull. Supply-side shocks may eventually cause inflation too, but those are seen, at least by most New Keynesians as being of secondary importance. Even Lucas, who has accepted for the most part the Real Business Cycle story, admits that one cannot explain the Great Depression and other such crises with real shocks.

Third, the MP rule implies that money is endogenous, and that central banks control the rate of interest and NOT the quantity of money. In other words, the old Monetarist rules about the rate of growth of money supply are out, since actual central banks very rarely have behaved in that way. This ‘innovation’ (not much of an invention since Wicksell used more than a century ago) within the mainstream took place without ever acknowledging the contributions of Kaldor (accomodationist tradition to endogenous money), Minsky (financial innovation tradition to endogenous money), Moore and other post-Keynesian authors.

Finally, it is rather clear that the NMC is essentially a neo-Wicksellian model (for a simple description go here), rather than Keynesian (New or otherwise). Not only the multiplier model was abandoned (and with it even the basis for fiscal policy activism, since the logic of Barro’s Ricardian Equivalence has been incorporated; see Wren-Lewis here), but also the concept of the natural rate that Keynes at least tried to get rid of has become central for policy analysis. And here is the Achilles’ heel of the NMC model.

Note that if the natural rate is not fixed, and in particular if it presents hysteresis or path dependence, and moves with the actual level of unemployment, then the basis of the NMC model falls apart. In other words, expanding demand might reduce the natural rate of unemployment and would not trigger inflation, and as a result would not require the central bank to hike the rate of interest to lean against the wind. That was, in a sense, the rationale for not hiking the rate of interest when unemployment fell below 6%, which many identified as the natural rate, during the Clinton boom. Greenspan suggested that productivity was going up (note that he didn’t necessarily say that productivity went up, and the natural rate down, as a result of the expansion of demand).

There are good logical reasons for not believing the natural rate story, as we know, associated to the limitations of the marginalist theory (see here). However, there are also reasonably well-established empirical problems with the natural rate hypothesis. The Real Business Cycles authors, in particular Nelson and Plosser in their classic paper (here), have long ago shown that output follows a random walk. In other words, changes in output are permanent, and there is no tendency for output to revert to its former trend following a shock, contradicting the natural rate hypothesis, or suggesting if one prefers that the natural rate moves with supply-side shocks and that the business cycle is the result of agents adjusting their behavior to the change in the natural rate.

As I noted before (here), the actual measure of productivity (Total Factor Productivity, TFP) used by RBC authors does NOT measure productivity, and most of their conclusions are irrelevant really. Also, as suggested above, it would be impossible to pin down the real shock that caused the Great Depression or the Great Recession, that have structural causes that are profound (in the patterns of consumption, private indebtedness and inequality) and that were triggered by financial shocks. However, the notion that the natural rate is not fixed, and that it changes significantly is actually quite important, since as we indicated, it suggests that policies that try to lean against the wind hiking the rate of interest when the economy is below the natural rate of unemployment are without foundation.

Heterodox authors would add to the RBC empirical observation about the fact that output is not mean reverting, that the supply or capacity limit of the economy is endogenously determined by autonomous spending (the supermultiplier that extends Keynes’ effective demand to explain potential output; for more go here). This does NOT mean that one can expand the economy without limits, since if the expansion of demand is faster than the movement of the capacity limit, eventually full employment would be reached and inflation (demand-pull inflation) might follow. Note, however, than since the 1930s in the US unemployment was below 4% (to say a relatively low number) only for four short periods, during the mid-40s, early 50s, late 60s and late 90s, with inflation occurring in the first three periods. Also, it suggests that the main barrier to the use of demand policies to achieve full employment, at least in developed countries with no balance of payments problems, is political. As Kalecki noted long ago, sound finance would be the political instrument to keep workers’ demands for higher wages in line. The NMC model is the modern incarnation of what Kalecki’s referred to as sound finance. So who is really surprised with the dominance of austerity policies?

Lars P. Syll: Shiller & Roubini's Fears of Swedish Housing Bubble Justified

By Lars P. Syll

The Swedish Riksbank has according to Lars E.O. Svensson been pursuing a policy during the last fifteen years that in reality has made inflation on average more than half a percentage units lower than the goal set by the Riksbank. The Phillips Curve he estimates shows that unemployment as a result of this overly “austere” inflation level has been almost 1% higher than if one had stuck to the set inflation goal of 2%. What Svensson is saying, without so many words, is that the Swedish Fed for no reason at all has made people unemployed. As a consequence of a faulty monetary policy the unemployment is considerably higher than it would have been if the Swedish Fed had done its job adequately. So far, so good — I have no problem with Svensson’s argument about the inadequacy of the Swedish inflation targeting policies. However, what makes the picture more complicated is that we do have a housing bubble in Sweden — it’s not just a figment of imagination the “bad guys” use to intimidate us with. [That said, I, of course, in no way want to imply that central bank interest rate targeting (and/or accommodations) is the best way to counteract housing bubbles. Far from it.]

Read the rest here.

Hysteresis and the natural rate

I've been teaching on the price and quantity interactions, and the natural rate or NAIRU (Non Accelerating Inflation Rate of Unemployment), that is the level of activity at which you have price stability. One of the papers assigned is the one by Franklin Serrano (here or here for a Spanish version; another assigned paper is this one by yours truly). By the way, I've dealt with the issue of hysteresis briefly before here, mostly to distinguish it from path dependency, following Setterfield (Serrano also suggests differences between heterodox and more conventional views on hysteresis).

As noted by Serrano, the research by Nelson and Plosser (1982) (here; subscription required) and Real Business Cycle (RBC) authors suggests that GDP follows a random walk, and as a result after a productivity shock (which they measure as changes in TFP, in spite of significant problems with that measure; see here) output does not return to its previous trend. The point is that once the output trend is affected there are persistent effects that change the trend itself, that is hysteresis. Fluctuations are variations of the optimal level itself.

Serrano correctly points out that "this means that the long run trend of output is not only partially determined by whatever drives short run output (presumably aggregate demand) but rather that potential output is actually fully determined by the trend of whatever drives actual output. As it is well known, this result of strong hysteresis in the output (GDP) series has been taken to provide evidence in favor of the 'real business cycles' strand of new classical macroeconomics in which the common element driving trend and cycle are factor supplies and their productivity." The natural rate or NAIRU is supply determined, but is variable (something that, in a different context, Robert Gordon would call the Time Varying NAIRU or TV-NAIRU).

Supply shocks imply that in a boom the potential output moves first, and actual output adjusts as individuals readjust to higher productivity. Hence, the output gap, if defined as the difference between actual and potential output, becomes negative. And if you believe in a Phillips curve and some sort of central bank monetary rule, a negative output gap suggests a deflationary pressure (and yes RBC authors do believe in endogenous money). Yes, that's what the RBC theory implies! In fact, according to Kydland and Prescott (1990): "the price level has displayed a clear countercyclical pattern."*

Serrano points out a simpler (Occam's Razor applies) explanation for the favorable evidence on hysteresis, namely that demand (in fact, the autonomous components of demand) determine potential output (the supermultiplier). Note that this approach does not require, as the RBC or the acceleracionist versions of the Phillips Curve, any definite relationship between prices and quantities. As noted here (and here) before, there is no reason to expect an unambiguous or systematic relation between prices and quantities, unless you think that prices are always driven by excess (or lack of) demand.

As Serrano argues the: "trend and the cycle indeed have a common nature as the empirical literature shows but this common nature reflects that both are explained by demand (not supply) factors" and "with full hysteresis in output levels and partial inertia on inflation, 'demand-pull' inflation is just a temporary phenomenon and therefore does not determine 'core' or persistent inflation." And it is hard not to agree on this with New Keynesians, and their dismissal of supply shocks as the main cause of business cycles. It is harder to agree with their insistence on a natural rate, even if it's variable.

* The obvious historical event they would have in mind is the stagflation of the 1970s. Note, however, that more often than not deflationary periods are contractionary, like the 1930s. Of course the oil shocks and the increase in costs can explain, together with wage resistance and price inertia, the inflationary pressures of the 1970s in a model that is perfectly compatible with demand driven recessions.

Phillips curve? What curve?

The Phillips curve is one of those 'regularities' that is more likely to exist in an economist mind than in reality. Figure below (from Ritschl and Straumann, 2010*) shows the 'trade-off' between inflation and unemployment in Europe in the 1920s and 1930.

They say quite candidly:

"the seemingly obvious connection between deflation and unemployment is less than easy to find in the data. ... Yet while there was ample variation in both unemployment and inflation during the interwar period, no systematic pattern seems to emerge in the data, even if the 1920s and the 1930s are looked at separately. The picture emerging from Figure [above] is rather that the natural rate of unemployment moved quite independently of inflation, irrespective of whether or not a country was on gold" (italics added).

Seemingly obvious, but inexistent in the data. In doubt economists choose the obvious and avoid reality. The other possibility that they do not seem to consider is that there is no natural rate at all, and no systematic relation between inflation and unemployment. Consider the data for the United States below (CPI change and unemployment data from Fred).

Note that there is also no clear correlation, unless you start assuming that it is unstable, and the curve is jumping, or that a supposed natural rate is jumping around too. Occam's Razor would suggest that if you have a theory that does not postulate any systematic relation between prices and quantities** it should be simpler and, hence, preferable.

* Ritschl, Albrecht and Straumann, Tobias (2010), "Business cycles and economic policy, 1914-1945." In Broadberry, Stephen and O'Rourke, Kevin H. , (eds.) Cambridge economic history of modern Europe. Cambridge University Press: Cambridge, UK , pp. 156-180.

** Note that classical-Keynesian theory implies that prices are determined by the technical conditions of production and exogenously determined distribution while unemployment derives from effective demand. Also, unemployment might have an indirect effect on inflation, through the effects of unemployment on the bargaining power of workers and, thus, on income distribution.

Prices and quantities

Mainstream economics suggests that prices and quantities should be treated simultaneously, as it should be if market prices as determined by supply and demand are at the center of the analytical framework. With supply and demand, it must be the case that the quantity produced, and the equilibrium price, both are determined simultaneously. Further, it is the case that, with price flexibility, the quantity produced is optimal from the perspective of the utilization of resources and the preferences of the agents. And if that is true for bananas, or any other commodity for that matter, it must be true too for labor and ‘capital.’

As I discussed before, in particular with respect to capital, this approach (supply and demand or marginalist), which contrasts with the surplus approach, has serious problems. Even if we dismiss, for simplicity sake, the subjective part (about preferences) and concentrate just on supply conditions, the difficulties are insurmountable. Producers only supply more at higher prices, which means that they must encounter increasing costs (diminishing returns). It cannot be the case that they are only willing to supply more at higher prices (higher remuneration) even if costs are not higher, since if one producer obtained higher remuneration free entry of new producers (attracted by the higher remuneration) would imply that more would be produced. So the supply curve depends on diminshing returns.

And diminishing returns are a highly improbable proposition, as Sraffa argued back in the 1920s. Not many producers, if any, would tell you that they don’t produce more because their costs would go up (and that would reduce demand as prices get higher). Most simply would reply that, although they could produce more at the same price, they don’t have enough demand. But the diminishing returns fetish dominates the profession (I won’t say anything here about imperfect competition, but Franklin Serrano noted here that this would be related to barriers to entry).

In the surplus approach, that concerns itself with the way in which societies reproduce themselves (usually in an amplified scale and with accumulation), prices and quantities are treated separately. Since costs of production, for a given technology, seem to be independent from the quantity produced (i.e. the cost would be the same if you produced slightly more or less), then the quantity can be taken as given for the discussion of the determination of prices. It is well know that, in that case, prices depend on the technology, which must allow for reproduction (including of the labor force) and by the way in which the surplus (what is left over, above and beyond the needs of reproduction) is divided between classes.

This separation between prices and quantities has been called by Garegnani the 'method of given quantities.' Note that it does not imply that the quantities are determined by supply or given at a level that is optimal from the point of view of utilization of resources. And that is why it is perfectly compatible with Keynes’ notion of a demand determined level of activity (yes you can, and should, add Sraffa to your Keynes, or Kalecki). Also, it does not mean that there is no technical change or that distribution does not affect output (all propositions that have been discussed here in the blog).

Be that as it may, I’m more interested in the macroeconomic implications of the mainstream views about prices and quantities. Since the late 1960s, these views have coalesced around the notion that, whereas there is a short run tradeoff between prices and quantities (the so-called Phillips Curve, PC), in the long run the tradeoff vanishes. Put simply, if you push demand too much (through fiscal and monetary policy), output would increase, unemployment fall (as per Okun’s Law: higher output implies lower unemployment) and inflation would accelerate. In the long run, however, the economy is self-adjusted and output cannot be above the optimal level, so the only effect of the expansionary policies would be inflation (Friedman dixit).

By the way, the same (the mainstream loves symmetry) is true for deflation. Yes it may cause some problems, but the system returns to full employment, even in the face of contractionary policies (the Greek should not worry about contraction, because markets would produce full employment if they are allowed to work; hence, the need of labor market flexibility). In the long run contractionary policies only would affect prices. In sum, supply and demand would lead to the optimal price and quantity, so if you get off my market (cranky old neoclassical guy would say), and stop expanding demand, there would be no inflation.

The evidence for a natural rate or for a PC is incredibly weak. In order to argue that there is a natural rate, the mainstream has basically suggested that it moves around all the time. So in the 1980s the natural rate was higher (in the US), when the actual unemployment was higher, than in the 1990s (particularly towards the end of the decade), when the actual rate was also lower. The ad hoc nature of the solution is evident. They tell you that the natural (which they measure as an average of the actual) is the attractor of the actual rate, and not the reverse.

The continuous ad hocery of the mainstream is on display now too. This is evident in discussion about the fears of inflation in the US, according to which the tripling of the monetary base would lead to hyperinflation (Krugman criticizes that here, and in several other places). Inflation does not take place, but it must be that expectations of inflation are low.

Also, it is evident in the IMF, the Bank for International Settlements (BIS) and other international institutions (since last year at least) arguments for fiscal and monetary contraction in the periphery (which is still growing faster than the center). The IMF believes fiscal expansion is no longer needed since “private demand has, for the most part, taken the baton” (IMF, 2011: xv). Moreover, the BIS argues that inflation is presently the main risk in an otherwise recovering world economy, and therefore suggests “policy [interest] rates should rise globally” (BIS, 2011: xii). According to this view then, if demand is controlled then prices would stop growing fast, and inflation would be under control.

Brazil actually did that last year, that is, fiscal contraction (with some monetary easing, but from a very tight stance, since it still maintains very high real interest rates). Output decelerated from 7.6% growth in 2010 to a mere 2.7% growth last year. A reduction of almost 5%. Did inflation then fell significantly as it should according to the mainstream? The Consumer Price Index (CPI) of the Fundação Getúlio Vargas increased from 6.24% to 6.36% in the same period, that is, it was almost constant. Something similar took place in Argentina between 2008 and 2009, when output decelerated around 10% (from high growth to negative growth), and inflation remained at high levels (fell perhaps around 6% or so, in an year that commodity prices collapsed). The mainstream argument is that the policies were not credible and as a result inflation expectations remained high.

In other words, it is not that the theory does not work, even though facts show that their predictions are false. It's a problem of the complexity of the phenomenon, and the need to account for expectations (anything can enter here, and no hypothesis can be tested if you take this seriously).

A simpler solution would be to admit that prices have little to do with demand (at most weaker demand and lower employment reduce the bargaining power of workers and reduce wage resistance). In other words, get rid of the notion of natural rate of unemployment and of a PC for which there is no reliable evidence (as should be done by any serious scientist). The contractions demanded by the IMF and the BIS led to recession (they affect quantities), and prices in Brazil did not fall because costs (which include imported goods and wages) did not fall. No need to suggest that expectations are responsible for inflation not falling. [Keen on his debate with Krugman has also referred to these subterfuges to keep theory in the face of lacking evidence; epicycles if you will]. By the way, expectations have been traditionally a way to argue that anything can happen, and allow theories that are not consistent with facts to get away with the incoherence. But I'll let that for another post.

References:

BIS (2011). 81st Annual Report. Basel, Bank for International Settlements, June.

IMF (2011). World Economic Outlook. Washington, DC, April.

It's the model stupid!

Brad DeLong, that together with Krugman has been a force for sanity within the mainstream, arguing for more fiscal expansion, shows why we need heterodox economists.  He says in a recent post:

This is a bad time to be an economist. If you were fresh from the womb and had no past opinions to defend, if you had never said anything notable before, it might be a fine time to be an economist. If you are one of those soap-opera characters who has complete amnesia and no memory of anything that they ever said or did or any intellectual position they took before January 1, 2010, it might be a fine time to be an economist. But for the rest of us--we who are now looking back at our opinions and analytic judgments and statements and pronouncements of the past 15 years and thinking: "how could I ever have been so stupid; how could I have missed so much?"--it is a bad time to be economist?

Four years ago we economists were writing learned papers about the "Great Moderation": about how it looked as though the governing institutions of the world economy had finally learned how to control and moderate if not completely eliminate the business cycle--the epileptic seizures of the economy that leave us with pointlessly high unemployment, pointlessly idle capacity, and pointlessly rusting away machines in spite of there being no fundamental cause for machines to be idle, factories closed, and workers unemployed.

Funny, I know of several economists that suggested that an economy based on debt-led consumption, on the basis of asset bubbles, was not sustainable and that a crisis was coming.  It would be tedious to cite all, and I'm lazy and don't want to find links to their papers, but a limited list of names (do a google search) would include Dean Baker, Jane D'Arista, Jerry Epstein, Jamie Galbraith, Wynne Godley, Thomas Palley, Bob Pollin and Lance Taylor (before hand I'm sorry for any significant omissions).  I knew enough to be sure that it was not sustainable.

But I'm glad that the best in the mainstream admit that it was stupid not to see it coming. I would suggest to him that part of the problem of the inability of the mainstream to see it coming is their theoretical framework.  The consensus macroeconomic model, based on an IS curve, a monetary policy rule, and a Phillips Curve with a natural rate of unemployment, in particular because it assumes that the economy automatically returns to the natural rate, is a flawed basis for understanding the real world.