I am now using Friday’s blog space to provide draft versions of the Modern Monetary Theory textbook that I am writing with my colleague and friend Randy Wray. We expect to complete the text during 2013 (to be ready in draft form for second semester teaching). Comments are always welcome. Remember this is a textbook aimed at undergraduate students and so the writing will be different from my usual blog free-for-all. Note also that the text I post is just the work I am doing by way of the first draft so the material posted will not represent the complete text. Further it will change once the two of us have edited it.
This is the continuation of the Chapter on unemployment and inflation – the series so far is:
- Unemployment and inflation – Part 1
- Unemployment and inflation – Part 2
- Unemployment and inflation – Part 3
- Unemployment and inflation – Part 4
- Unemployment and inflation – Part 5
- Unemployment and Inflation – Part 6
I am now continuing Section 12.6 on the Phillips Curve …
Chapter 12 – Unemployment and Inflation
MATERIAL HERE NOT REPEATED[PICKING UP FROM THIS POINT …]
Consider Figure 12.8, which shows the combinations of the unemployment rate and the annual inflation for the US economy from 1948 to 1980. The blue diamonds show the observations for the period 1948 to 1969, and the black curve is the logarithmic regression between the inflation rate and the unemployment rate. So it shows a simple price Phillips curve of the type depicted in Equation 12.3.
The curve fits the data quite well and shows the typical trade-off between unemployment and inflation, that was considered to be fit for exploitation by policy makers, intent on keeping unemployment close to full employment.
However, consider the observations depicted by the red squares, which cover the period from 1970 to 1980. Those observations are clearly inconsistent with the stable Phillips curve representation and seem to suggest a positively sloping relationship.
This apparent shift in the Phillips curve was cast as a “collapse” in the relationship and led to accusations that the underlying conceptualisation of the Keynesian Phillips curve was flawed.
Source: US Bureau of Labor Statistics. The unemployment rate series begins in 1948. The inflation rate is measured as the annual rise in the Consumer Price Index.[NEW MATERIAL TODAY STARTS HERE]
Consider Figure 12.9, which shows the relationship between the US unemployment rate (horizontal axis) and the annual inflation rate for the period 1948 to 2012, subdivided into three periods: 1948-1969; 1970-1980; and 1981-2012. As before the lines depict the logarithmic regression between the inflation rate and the unemployment rate for the three sub-periods.
What was considered to be a relatively stable relationship in the 1950s and 1960s, became unstable in the 1970s and suggested a positive relationship between inflation and unemployment. By the 1980s, as inflation moderated, it became hard to determine any relationship between inflation and unemployment in the US economy.
From an empirical perspective, then, the faith that the Phillips curve was a stable relationship that could be exploited in some predictable fashion by polict makers to maintain low unemployment became highly questionable.
Source: see Figure 12.8
We always have to be very careful when we visualise data in this way. The observations between 1970 and 1980, may in fact be signifying a shifting Phillips curve relationship and the regression line is just picking up the shifting function. In other words, the relationship is unstable over time although it may permit a short-run trade-off between the inflation rate and the unemployment rate.
That conjecture is represented in Figure 12.10, which shows stylised Phillips Curve for the US data from 1948-2012. Please note that these are not actual estimated Phillips curves. They just demonstrate that there might be a family of such curves and each one might remain in place for a short-period, which woudl provide policy makers with a false sense of security. At some point the curve might shift out to a new loci and inflation would accelerate.
Source: see Figure 12.8
It should be clear from our earlier discussion of the work of A.J. Brown that the assumption that the Phillips curve was stable was always problematic.
It was known that the Phillips curve was susceptible to a sudden and/or large increase in inflation. The econometrically-estimated consumption functions in the large macroeconomic policy models, which were popular in the 1960s also became unstable in the 1970s. Some economists successfully showed that the failure of the large-scale econometric models to forecast such variables such as savings and consumption in the early 1970s could be traced to the misspecification of the structural consumption function. Most of these models ignored the possibility that rising inflation would influence consumption (for example, if consumers expect prices to rise quickly in the future they may bring forward consumption decisions).
The breakdown of the Phillips curve in the late 1960s was another “econometric” function that was misspecified because it also ignored the possibility that rising inflation might become self-fulfilling as workers and firms sought to protect their real wages and real profit margins.
Another consideration as to why the discussions about instability were largely ignored is that the “textbook” model of the Phillips curve was very attractive in its simplicity. Textbooks typically stylise discussions and eschew complicated stories for the sake of pedagogy. You will be aware that we have not taken this approach in this text. We consider a rich treatment of institutions and history to be an important part of the learning process in macroeconomics.
The work of A.J. Brown and others was insightful and probably too deeply grounded in the institutional literature to be acceptable for the way textbook writers chose to advance their pedagogy at the time.
It is probable that had Brown’s work on instability and the way changes in the institutions of wage and price determination (for example, wage bargaining and real wage resistance) change the trade-off between inflation and unemployment been more recognised, the subsequent history of the Phillips curve, which we discuss next, might have been very different.
It is a fact that the mainstream Keynesian consensus in the 1960s abstracted from the potential instability that was rooted in the institutional nature of wage and price setting. Instead, policy makers pursued the attractive notion that they could permanently maintain low unemployment rates as long as they ensured effective demand was sufficient relative to the non-government sector’s saving plans and any demand leakages from net exports.
However, the legacy of the Keynes and Classics debate persisted through the 1950s and 1960s. The neo-classical school was unwilling to accept the basic insights provided by Keynes that effective demand drove output and national income and that the capitalist monetary system was susceptible to crises of over-production.
In the late 1960s, this on-going debate about the effectiveness of fiscal and monetary policy in counter-stabilising the economic cycle was rehearsed within the Phillips curve framework.
A group of economists, centred at the University of Chicago were opposed to government attempts to maintain full employment. Their argument largely reflected their belief that a self-regulating free market would generate optimal outcomes. In other words, they were adherents of the competitive neo-classical model and considered most government intervention to be problematic.
This opposition at the microeconomic level manifested in demands for widespread deregulation in product, labour and financial markets and a major retrenchment in the size of government.
At the macroeconomic level, the Phillips curve became the “battleground”. Policy makers during the “stable” Phillips curve policy era assumed they could permanently target a low unemployment rate and incur some finite inflation rate as a consequent. The extent to which inflation rose was determined by the steepness of the Phillips curve, which was considered to be relatively flat.
The emerging Monetarists, who eschewed government intervention, challenged that view and asserted that there was no permanent (long-run) trade-off between inflation and unemployment. They claimed that, ultimately the market would ensure the unemployment rate was stable around its so-called natural rate and attempts by government to push it below this rate would lead to accelerating inflation.
Chicago economist Milton Friedman was the most vocal Monetarist and in a famous article in 1968 outlined what became known as the accelerationist hypothesis[Full reference: Friedman, M. (1968) ‘The Role of Monetary Policy’, The American Economic Review, 58(10), March, 1-17]
12.X The Accelerationist hypothesis and the Expectations-augmented Phillips curve
The empirical instability in the relationship between unemployment and inflation opened the way for what became known as the Monetarist paradigm in macroeconomics to gain ascendancy.
The Monetarists reinterpreted the trade-off by adding the role of inflationary expectations, and in doing so, revived the Classical (pre-Keynesian) notion of a natural unemployment rate (defined as equivalent to full employment),. The devastating consequence of this assertion was the rejection of a role for demand management policies to limit unemployment to its frictional component.
The Phillips curve also became a tool in the hands of the Monetarists to regain the ground they had lost to the Keynesians. With the support of the textbooks the model endured even though the original model was lost in the process.
Despite the rich Keynesian history discussed earlier, it was easy for Friedman and others to hijack the debate. The Keynesians were operating in a dichotomised framework – at the macroeconomic level they had adopted the Phillips curve, yet they had developed very little theory to underpin it. They were particularly tense and uneasy about what were referred to as the microeconomic foundations of the relation.
There were attempts by the Keynesians to justify the Phillips curve as a competitive adjustment process, such that if there was growing demand for labour wages rose as unemployment fell.
But the Monetarists claimed that in a competitive market workers cared about real wages rather than nominal wages and so the basic Phillips curve which only focused on the relationship between percentage change in money wages and the unemployment rate was defective.
The accelerationist hypothesis was advanced in 1968 by Milton Friedman before the empirical breakdown of the relationship between inflation and unemployment emerged in the early 1970s.
So while the Phillips curve presented the Monetarists with the opportunity to debate the failings of the mainstream Keynesian analysis, it was the empirical havoc created by the 1970s oil price shocks which added weight to their (flawed) arguments.
It is important to note, that nothing had really changed in the modern statement of Monetarism that had not already been shown to be deficient, albeit in different terms, by Keynes and others.
As noted above, the Phillips curve was just one of a number of macroeconomic equations in the large policy models that ignored inflationary expectations. The misspecification was not significant while inflation was negligible. Once inflation rates soared throughout the world with the oil price rises of the early 1970s, all these misspecified relations broke down.
The Monetarists used that empirical chaos to cast the theoretical ideas of Keynesianism into disrepute. Monetarist thought emerged from this wreckage as being eminently plausible. It was a serendipitous period for the Neoclassical economists because they managed to reassert the issue of real wage bargaining before the empirical relations broke down.
Although in the mid-1960s, the Monetarist theoretical structure had undergone harsh criticism from economists like Robert Clower and Axel Leijonhufvud, the empirical shift in the Phillips curve in the early 1970s was interpreted as a validation of the Monetarist concept of a natural rate of unemployment and the negative connotations for aggregate demand management that this concept invoked.
There were two basic propositions that Friedman asserted in his attack on the Phillips curve.
First, that there is a natural rate of unemployment, which is determined by the underlying structure of the labour market and the rate of capital formation and productivity growth. For example, he claimed that minimum wages and welfare support increased the natural rate. The natural
Friedman (1968: Page 8 ) wrote:
At any moment of time there is some level of unemployment which has the property that it is consistent with equilibrium in the structure of real wages. At that level of unemployment, real wage rates are tending on average to rise at a ”normal” secular rate, i.e., at a rate that can be indefinitely maintained so long as capital formation, technological improvements, etc., remain on their long-run trends. A lower level of unemployment is an indication that there is excess demand for labor that will produce upward pressure on real wage rates…. The “natural rate of unemployment”, in other words, is the level that would be ground out by the Walrasian system of general equilibrium equations, provided there is imbedded in them the actual structural characteristics of the labor and commodity markets, including market imperfections, stochastic variability in demand and supplies, the cost of gathering information about job vacancies and labor availabilities, the costs of mobility, and so on.
He later noted (Friedman, 1968: 9) that the natural rate of unemployment is not “immutable and unchangeable” but reflects his belief that it is a real variable, which is insensitive to monetary forces. That is, he considered increasing nominal aggregate demand would not reduce the natural rate.
He also considered that the natural rate was, in part, “man-made and policy-made”. For example, Monetarists argued that imposing minimum wages and providing unemployment benefits would increase the natural rate.
Second, that the Phillips curve is, at best, a short-run relationship that can only be exploited as long as workers suffer from money illusion and confuse money wage rises with real wage rises. In other words, any given short-run Phillips curve is dependent on workers assuming that price inflation is stable. Under that assumption, there is no role for price expectations in the Phillips curve.
However, Friedman and others argued that eventually workers would realise that their real wage was being eroded in price infation outstripped money wages growth. In doing so, they would start to form expectations of continuing inflation.
As a consequence, workers would build these inflationary expectations into their future outlook and pursue real wage increases, which reflected not only the state of the labour market (relative strength of demand and supply) but also how much they expected prices to rise in the period governed by the wage bargain.
The Monetarists argued that if the government attempted to reduce unemployment below the natural rate, then as the inflation rate rose, workers would demand even higher money wages growth to achieve their desired real wage levels. Ultimately, all that would result was an accelerating price level.
Figure 12.11 captures the accelerationist hypothesis. The short-run Phillips curves are shown conditional on a specific expectation of inflation (in this case, zero inflation, 3 per cent and 6 per cent).
Start at Point A, where there is zero inflation. The government forms the view that the unemployment rate is too high and adopt a lower target rate of unemployment, UT.
The government stimulates nominal aggregate demand to push the economy to point B along the SRPC Pdot=0. The increased demand for labour pushes up both the inflation rate (to 3 per cent) and money wage rates in the labour market. Monetarists assumed that the price level accelerated more quickly than money wages because the only way firms would employ more workers would be if the real wage had fallen. The Monetarists resurrected the Classical labour market and placed it at the centre of their attack on Keynesian macroeconomics.
Why would workers supply more labour if the real wage was falling? In the Classical labour market it is assumed that labour supply is a positive function of the real wage so workers will withdraw labour if the real wage falls.
The Monetarist sleight of hand was to impose what they considered to be asymmetric expectations on the workers and firms. The firms were assumed to have complete price and wage information at all times so they knew what the actual real wage was doing at any point in time. However, the workers were assumed to gather information about the inflation rate in a lagged or adaptive fashion.
Thus, workers are assumed to be initially oblivious to the rising inflation (they have zero inflationary expectations) and, thus, they mistake the rising nominal wages for an increasing real wage. As a consequence, workers willingly supply more labour even though the real wage has actually fallen.
To understand the dynamics on the demand and supply sides of the Classical labour market, please refer back to the discussion in Chapter 11 Keynes and the Classics.
The basic point is that workers care about real wages not money wages but take some time to differentiate between the two.
But the Monetarists asserted that Point B is unstable.[TO BE CONTINUED]
NEXT WEEK – THE HYSTERESIS CHALLENGE TO THE ACCELERATIONIST HYPOTHESIS AND THE ROLE OF BUFFER STOCKS IN THE INFLATION PROCESS (YES … the Job Guarantee enters the picture!).
The Saturday Quiz will be back again tomorrow. It will be of an appropriate order of difficulty (-:
That is enough for today!
(c) Copyright 2013 Bill Mitchell. All Rights Reserved.