Unemployment and inflation – Part 3

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:

I am now continuing Section 12.6 on the Phillips Curve …

Chapter 12 – Unemployment and Inflation


12.6 The Phillips curve

In Chapter 9, we derived what we termed to be a General Aggregate Supply Function (Figure 9.5) and the reverse-L shape. The horizontal segment was explained by the price mark-up rule and the assumption of constant unit costs. In other words, firms in aggregate are assumed to supply as much real output (goods and services) as is demanded at the current price level set up to some capacity limit.

Figure 9.5 became vertical after full employment because beyond that point the economy exhausts its capacity to expand short-run output due to shortages of labour and capital equipment. At that point, firms will be trying to outbid each other for the already fully employed labour resources and in doing so would drive money wages up.

Under normal circumstances, the economy will rarely approach the output level (Y*) which means that for normally encountered utilisation rates the economy faces constant costs.

We acknowledged in Chapter 9, however, that rising costs might be encountered given that all firms are unlikely to hit full capacity simultaneously. Ww noted that the reverse-L shape simplifies the analysis somewhat by assuming that the capacity constraint is reached by all firms in all sectors at the same time.

In reality, bottlenecks in production are likely to occur in some sectors before others and so cost pressures will begin to mount before the overall full capacity output is reached.

This recognition leads us to consider the so-called Phillips curve, which was originally constructed in the work of A.W. Phillips published in 1958, as a relationship between the percentage growth in money wages and the unemployment rate. Later, economists constructed the relationship as being between the general inflation rate and the unemployment rate.

In the pre-Keynesian era, unemployment was considered to be a voluntary state and full employment was thus defined in terms of the employment level determined by the intersection of labour demand and labour supply. So by construction, full employment reflected the optimal outcome of maximising, rational and voluntary decision making by workers and firms. At the so-called full employment real wage, any worker wanting work could find an employer willing to offer the desired hours of employment and any employer could fill their desired offer of hours from the services of willing employees.

In the Section 12.4, we discussed the fact that the trade-off between inflation and unemployment had been a subject of discussion since the time of the Classical economists, but it never had a prominent place in the debate and was dominated by the evolution of the Quantity Theory of Money.


Early Keynesian empirical work on inflation and unemployent

In the 1930s, Dutch economist Jan Tinbergen published the first econometric study of the trade-off between inflation and unemployment in 1936. Tinbergen became famous for his 1939 League of Nations project which attempted to provide empirical justification for the emerging Keynesian view that governments should intervene to stabilise business cycle fluctuations (Reference: Tinbergen, 1939).

Tinbergen was the first Phillips curve if we take that to mean a model with causality running from excess demand in the labour market to wage changes. The model was thus based on price adjustment reacting to quantity disequilibrium in the labour market with no presumption of full employment. Given that excess demand is not directly observable, Tinbergen used employment relative to its trend level as the excess demand proxy.

Tinbergen also included a price change term on the right-hand side of his equation explaining money wage changes which was lagged one period. This was to represent catch-up behaviour or cost of living adjustments to nominal wages. In other words, Tinbergen had a model of wage inflation dependent on excess labour market demand and a shift parameter (in his case the lagged inflation term). The estimated model clearly implied a trade-off between wage inflation and the state of the labour market.

In the 1940s and 1950s, the US Cowles Commission were the first to construct large-scale macroeconometric models, which were designed to aid policy makers in determining the potential consequences of their fiscal policy choices. A key figure in this work was American economist Lawrence Klein, who acknowledged the influence of Jan Tinbergen in designing models of wage inflation.

Klein’s wage inflation equation used unemployment as the (inverse) proxy for excess demand for labour and included price change variables on the right hand side to capture real wage aspirations of workers. So workers would try to gain money wage increases to match inflation but unemployment would temper their capacity to succeed in their bargaining aims. (Reference: Klein (1985)).

From an historical perspective, the work of the Cowles Commission in developing models of inflation (Phillips curve) and the relationship between unemployment and GDP gaps (Okun’s Law) was the foundation stone for the 1960s Keynesians and became the centrepiece of macroeconomic orthodoxy during this period in place of the Quantity Theory of Money, which had been thoroughly discredited by Keynes himself and subsequent empirical research.

It is important to acknowledge that the work of Klein and his team showed they knew that the economy could undergo nominal wage inflation independent of the state of the labour market if bargaining agents formed expectations that the inflation rate was rising.

They considered workers formed expectations of future inflation which influenced the way they bargained for wage rises with their employers. They explicitly modelled the change in money wages as a function of lagged inflation because they considered it took time for real wage aspirations to feed into bargained outcomes.

As we will see later, the Monetarists in the late 1960s claimed that it was Milton Friedman and Edmund Phelps who showed that government aims to reduce the unemployment rate using expansionary fiscal and monetary policy would be futile because of the role of expectations of inflation. The reality is that the economists long before Friedman and Phelps clearly understood that workers would consider expected inflation when bargaining for money wages.

The point to understand is that following the publication of the General Theory in 1936, Klein and his cohort advanced the Keynesian paradigm by giving it empirical authority. Their efforts in estimating the inflation-unemployment nexus was empirically based and they did not provide a strong theoretical basis for their models.

Arthur Brown’s The Great Inflation

In the 1950s, the Phillips curve became a centrepiece of macroeconomic analysis. The first major contribution (prior to Phillips’ own 1958 publication) was provided by the English economist Arthur Brown in his book – The Great Inflation, 1939-51.

[Reference: Brown, A.J (1955) The Great Inflation, 1939-51, Oxford University Press: Oxford]

There was an interesting – obituary – written by Tony Thirlwall on the event of Arthur Brown’s death in 2003.

Thirlwall wrote that Arthur Brown:

… precisely anticipated the Phillips curve, which plots the inverse relation between wage and price inflation and the rate of unemployment.

He never received the credit he deserved for this empirical finding; strictly speaking, the Phillips curve should be called the Brown curve

Brown, who was firmly in the Keynesian mould, provided an account of the role of expectations and real wages in the determination of the trade-off between inflation and unemployment, thus adding theoretical substance to the empirical work of Tinbergen and Klein et al.

Brown was also the first theorist to investigate the question of the instability of the wage change-unemployment relationship, a subject that was at the centre of the Monetarist attack on the Keynesian orthodoxy in the late 1960, the legacy of which has influenced macroeconomic theory and policy design ever since.

Brown’s work also outlined the relationship between the price-wage spiral mechanism, which can drive inflation and the distributional struggle between labour and capital over the available real income. In this sense, he anticipated a competing claims explanation of inflation, which became popular in the 1970s among Post-Keynesians and Marxists. We will discuss these conceptions of inflation later in this Chapter.

His aim was to explain the underlying causes of the inflation and he said that merely characterising this as (Brown, 1955: 16):

… a propensity of the community to spend more than its current income … does not throw much light upon the causes of inflation.

He believed that the “reason for this propensity” (Brown, 1955: 17):

… may go deep into the market institutions and the social and political structures of the communities concerned …

Thus, he was not satisfied with an approach that was ahistorical such as the Quantity Theory of Money. To understand the inflationary process, Brown thought it was essential to understand social, economic and political factors and their interactions.

Brown assumes an open economy where firms have price setting power is used and fixed nominal contracts between workers and employers are common. Thus he was not convinced that the competitive models of firm behaviour that neo-classical economics promoted were of any use in explaining what happens in a capitalist economy where firms have market power exists.

He also recognised that firms do not adjust prices in response to changes in demand for their goods and services because there are significant costs of price adjustment. He wrote (Brown, 1955: 80) that prices are:

… fixed by producers in relation to costs of production (which depend on factor prices), and for wages to be fixed either automatically or by bargaining in relation to the prices of consumers’ goods.

In other words, there is a wage-price dependency, which underpins the inflation mechanism.

Brown (1955: 100-101) produced two “Phillips curve” scatter plots using UK data which showed the relationship between the rate of unemployment (on the horizontal axis) and the percentage change in wage rates (on the vertical axis), in one case, and the percentage change in hourly earnings (1920-51), in the second diagram (1920-1948).

Figure 11.3 depicts a stylised version of the relationship between unemployment and money wage inflation that was posited by Brown.

Figure 11.3 The basic Phillips curve

The causality underlying the scatter plots was Keynesian such that the business cycle was driven by fluctuations in effective demand. In times of strong growth, the labour market disequilibrium (excess demand for labour) increases bargaining power of unions and reduces unemployment and this leads to an increase in the rate of money wages growth. Conversely, when the expansion ceases and unemployment starts to rise again, the rate of growth in money wages declines.

Brown also thought the relationship between wage changes and the unemployment rate was non-linear. Thus, wage changes are larger when the unemployment rate is low than when the unemployment rate is high. The curve implied is very flat at high unemployment rates and then becomes very steep, if not vertical, at low unemployment rates.

Brown considered that attitudes to the distribution of real income between wages and profits were also influential in the bargaining process. Workers will resist attempts by firms to increase profit margins and cut real wages through price rises. Firms, also use their price setting power to try to gain a greater share of real income. They also try to pass on the burden of rising raw material prices to workers.

The conflict over the distribution of real income can thus lead to a wage-price spiral developing where the respective bargaining strengths of labour and capital perpetuate an inflationary process as each defend their real claims on income.

Brown (1955: 105) wrote that the development of a wage-price spiral is dependent on the:

… aims of the two parties who are competing for the real income of the country or their success in achieving those aims.

This idea became the centrepiece of what we now refer to as the “competing claims” theories of inflation, which depict the inflation process as resulting from incompatible claims on total nominal income by workers and firms, which exceed the total available.

Workers negotiate real wage targets via money wage demands on firms, who in turn pursue some target markup (as a vehicle for a desired rate of return). Prices may be slow to adjust in a time of rising costs due to the costs of price adjustment. If the sum of the claims exceeds national income, either or both parties may use their price-setting power to achieve their targets. Inflation is the outcome of the distributional conflict.

It is thus important to acknowledge that while Brown considered the inflationary process was conditioned by the state of the economy, he also explicitly recognised that inflationary expectations could drive the wage-price spiral independently of excess demand.

Once prices have been given a sufficiently strong upward push by any cause whatever, it is possible that further increases may be expected, and the automatic fulfillment of this expectation, in advance of the time at which the further increase was anticipated, may then cause a further upward revision of prices expected to rule in the future. A mechanism of this kind can, clearly, produce an indefinitely accelerating price increase, provided that certain requirements are met.

The conditions outlined by Brown (1955: 5-6) whereby expectations of inflation can drive price increases independent of excess demand factors are:

  • Prices have to be flexible enough to change when expectations change.
  • Money supply growth must be sufficient to match price inflation or else real demand falls.

The speed of adjustment of price changes to changes in price expectations does not rely on perfect flexibility. It can take some time for expectations to work their way through the price change process without negating their influence.

A final point relating to Brown’s work relates to his recognition that the relationship between wage changes and the unemployment rate was not stable. He argued convincingly that the relationship went through several discrete periods – Pre World War I, between the wars, then Post World War II.

He considered historical epochs to offer quite different bargaining conditions and price setting power. His contribution teaches us that universal models of economic behaviour always have to be tempered by a contextual and historical understanding.

The work that followed Brown, however, downplayed the recognition that the relationship between wage changes and unemployment might not be stable over time. As we will see, this shift in emphasis had significant historical ramifications in the development of thinking about inflation and unemployment and subsequent policy design.

Paul Sultan (1957) – “I discovered the Phillips curve”!

While Arthur Brown went within one stroke of producing the graphical Phillips curve of the type depicted in Figure 11.3, it was the American text-book writer, Paul Sultan who published the first textbook version of what is now known widely as the Phillips curve (that is, Figure 11.3).

[Reference: Sultan, P. (1957)]

Sultan’s graph showed the annual percentage change in the price level on the vertical axis and the rate of unemployment on the horizontal axis. Importantly, this was not the relationship that Phillips published in 1958. His was in terms of money wage changes rather than price inflation.

Sultan wrote (1957: 555) that:

… the line relating unemployment to inflation … [reference to his Figure 24] … is strictly hypothetical, but it suggests that the tighter the employment situation the greater the hazard of inflation. …Assuming that a fairly precise functional relationship exists between inflation and the level of employment, it is possible to determine the ‘safe’ degree of full employment. In our hypothetical case, we are assuming that when unemployment is less than 2 per cent of the work force, we face the dangers of inflation. And when unemployment is larger than 6 per cent, we face the problem of serious deflation.

Sultan did not consider the role of expectations and the question of stability.

Phillips work on inflation

In 1958, New Zealand economist Bill Phillips published a statistical study, which showed the relationship between the unemployment rate and the proportionate rate of change in money-wage rates for the United Kingdom. He studied that relationship for the period 1861 to 1957.

He had earlier written a paper in 1954 which can be considered a precursor to his 1958 empirical study.

[REFERENCE: Phillips, A.W. (1958) 'The Relation between Unemployment and the Rate of Change of Money Wage Rates in the United Kingdom, 1861-1957', Economica, 25(100), 283-299. The link is if you have a subscription at your library to JSTOR.]

Phillips believed that given that money wage costs represent a high proportion of total costs that movements in money wage rates will drive movements in the general price level.

The Phillips curve has been used by macroeconomists to link the level of economic activity to the movement in the price level. In the Phillips curve framework, the level of economic activity is represented by the unemployment rate. The presumption is that when the unemployment rate rises above some irreducible minimum then economic activity is declining and as the unemployment rate moves towards that irreducible minimum the economy moves closer to full capacity and full employment.

We will see in a later section that the Phillips curve and Okun’s Law, which links changes in the unemployment rate to output gaps (the difference between potential output and actual output) coexist comfortably in macroeconomic theory. The latter provides the extra link between unemployment and output.

In some textbooks you will find inflation models that conflate the two concepts (Phillips curve and Okuns’ Law) and directly relate the inflation rate to the output gap. We prefer for reasons that will be obvious not to take that approach in this text book.




Saturday Quiz

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.

This entry was posted in MMT Textbook. Bookmark the permalink.

4 Responses to Unemployment and inflation – Part 3

  1. AmEconist says:

    A very informative synopsis.

    I would understand that the vertical line in Figure 11.3 is
    “Change in money wages (%)/100 = (Average Wage1 – Average Wage2) / Average Wage”
    where Average Wage1 and Average Wage2 are measurements taken in subsequent time periods.

    Unemployment Rate would be the familiar point-in-time measurement of unemployment rate.

    I think of what happened in Alaska during the Alaska Gold Rush. Micro-inflation caused by speculative demand. Maybe this example holds some clues for why the money supply increase today seems to be doing less than expected to boost prices or demand.

  2. Tim Watson says:

    Hi Bill,

    I would also include a discussion on transient supply shocks in the Phillips curve equation. I note that Phillips’ 1959 paper that he wrote at Melbourne university included a role for both import and export prices.

    Phillips noted that import prices were likely to have an impact on wage inflation through their impact on consumer prices. Modern versions capture this immediate effect, and also second round pass through effects via imported inputs in the local production chain.

    Phillips also believed large increases in export prices were likely to have an additional effect on wage rates because this tended to increase the earnings of agriculturalists which in turn resulted in higher arbitration court awards granted to workers. Of course most modern expositions leave out export price shocks, but still feature a role for import, energy or other commodity prices.

    I would also talk about early references to the importance of the rate of change of unemployment, as well as the level of unemployment, and how this ties in with ideas concerning hysteresis. Phillips posited that this was important in his 1958 paper, and the relationship between inflation and the rate of change of unemployment was formalised by Lipsey in 1960.

    Enjoying reading your textbook instalments, and look forward to its completion.


  3. “Figure 9.5 became vertical after full employment because beyond that point the economy exhausts its capacity to expand short-run output due to shortages of labour and capital equipment.”

    It’s not just labour and capital equipment that run into short supply. It’s also materials, services, intermediate goods and so on. The FUNDAMENTAL item that is in short supply when an economy is at capacity is LABOUR: reason is that a majority of the cost of anything can be traced back to the cost of labour (with interest and profit accounting for the rest).

    I.e. I suggest you just say “due to shortages of labour”. But if you want to drag capital equipment into it, then by the same token you need to drag materials, intermediate goods, services, etc etc into it.

  4. Esp Ghia says:

    Great point Tim Watson. Hey, you’re not from Dimboola, are you?

Leave a Reply

Your email address will not be published. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

To prove you're a person (not a spam script), type the answer to the math equation shown in the picture. Click on the picture to hear an audio file of the equation.
Click to hear an audio file of the anti-spam equation