Part three – a value theory of inflation
In this post, I return to my review of the theories of the causes of inflation in modern economies. In the first part, I discussed the mainstream theories; in the second part, I discussed heterodox and other Marxist theories. In this third part, I present a value theory of inflation developed by Guglielmo Carchedi and myself.
Part One is here, Part two here
In our view, any explanation of inflation in modern capitalist economies must start with Marx’s value theory. In all the conventional, heterodox and even most Marxist-claimed theories, the role of value creation is missing. This creates a black hole in any analysis of inflation. For Marx, value is the expenditure of human labour power in the abstract, and is measured in labour time, i.e. the time worked by labour for capital. This is the time actually worked for the production of commodities. Not all abstract labour is productive of value: some labour is unproductive (basically commercial, financial and labour in armaments and real estate). Unproductive labour merely redistributes value. Productive labour is that which creates new value and surplus value. So, in our value theory of inflation, value is measured by the hours worked in the productive sectors.
The realisation of value requires the exchange of the different commodities and thus money acts as a medium of exchange. Money makes exchange possible and so represents value. If money represents value, then changes in money are determined by changes in value. This is a key difference between the monetarist and the value theory of inflation, as I explained in part one. The monetary authorities only react to value changes either by increasing or decreasing the supply of money or raising or reducing interest rates on holding cash or borrowing; but this does not change value.
In mainstream economics, inflation is defined as the change in the prices of commodities on the market. Instead, in our value theory of inflation, inflation is the difference between changes in money in circulation and changes in value production. This difference will become important in considering the impact on workers’ wages.
We define and measure value as the hours of productive labour worked in any period. We define and measure money supply as money in circulation. Not all money supply in a capitalist economy circulates for the purposes of purchases of goods and services. Some portion will be hoarded i.e held in reserve by capitalists. Another portion will be used to purchase financial assets (bonds, stocks etc) and real estate (land, buildings). This money will not circulate to purchase goods and services and so will not affect any change in the prices of goods and services consumed by workers.
So in our measure of money supply we adjust money supply (deposits in banks) by the size of bank money reserves held at the central bank and by the velocity of money in any period. The velocity of money is a measure of the turnover in the money supply. If the velocity is greater than one, it means that the money supply has been used more than once in any period and so adds to total money in circulation. If the velocity of money is below one, it signals that the holders of money are hoarding more of the supply rather than putting it into market for goods and services. Accounting for these adjustments, we obtain a measure of ‘money in circulation’ for any period.
We can measure the size of hoarding and financial speculation by the difference in the rate of change in money supply and the rate of change when adjusted for hoarding and speculation in financial assets. In the second half of the 20th century, there was little difference in the rate of change in money supply (bank deposits etc, or M2 in US financial statistics) and the adjusted money in circulation in the US. Indeed, M2 adjusted for hoarding and speculation generally rose faster than M2. That’s because the velocity of money was usually greater than one and hoarding and financial speculation was minimal. However, after 2000 and especially after 2008 and through the 2010s, the gap between money supply growth and the growth of money in circulation rose sharply. This was the period of so-called ‘quantitative easing’ adopted by the US Federal Reserve. Much of the Fed’s monetary injections mostly ended up being hoarded or used for speculation in financial assets rather than for the purchases of goods and services. US CPI inflation fell to virtually zero during this period, while financial asset prices soared.
Source: FRED, author’s calculations
To obtain the value rate of inflation, we measure the difference in the change in adjusted money supply against the change in hours worked in the productive sectors of an economy. If there is no difference, there will be no inflation. Inflation emerges when the percentage change in money in circulation is greater than the change in value created. Deflation emerges when the rate of change in money circulation falls below the rate of value growth. Inflation rises when the difference widens between the rate of change of money in circulation and the rate of change in hours worked, and when the difference narrows, inflation falls (disinflation).
What determines these two factors and which of the two is the determining and determined factor? Let us begin with productive hours worked, our measure of value created. Over the whole period analysed (1949-2022), hours worked increased in the productive sectors of the US economy. That’s because the increase in the number of workers employed rose more than sufficently to compensate for any fall in hours worked per employee, except when workers were laid off in recessions (1957-8, 1974-5, 1980-2, 1991, 2001, 2008-9 and 2020). Then overall hours worked fell.
Source: FRED, authors’ calculations.
In Marxist theory, investment growth leads to an increase in the ratio of constant capital (means of production) relative to labour employed ie a rise in the ‘organic composition of capital’. The reciprocal of a rising organic composition is a decrease in value created per unit of capital invested, in other words, a fall in the rate of profit on capital invested. Over the whole period 1949-2022, there is a positive correlation between the fall in the rate of profit on capital and slowing growth in hours worked relative to capital invested.
Source: FRED, author’s calculations
The annual ‘value rate of inflation’ is the difference between the annual change in the adjusted money supply and the annual change in productive hours worked. Over the whole period 1949-2022, adjusted money supply growth averaged 6.6% a year and growth in hours worked in the productive sectors averaged 1.4% a year. So the value rate of inflation averaged 5.2% a year (6.6%-1.4%).
Source: FRED, author’s calculations
Over the whole period, the rate of money supply growth falls, while the change in hours worked rises (slowly). So the value rate of inflation falls over the whole period – in effect, there was disinflation (a falling inflation rate). Disinflation has been the long-term trend since the end of WWII until the end of the 2010s.
Source: FRED, authors’ calculations
But we can discern two sub-periods within 1949-2022. The first is from 1949-81 and the second is from 1982-2019. In the first period, the annual value rate of inflation rises, constituting an inflationary period. In the second period, the annual value rate of inflation falls, constituting a disinflationary period. The value rate accelerates in the first period because the adjusted money supply grows faster than hours worked (see Figure above). This was particularly the case in the 1970s, when hours worked stagnated or fell (see the hours Figure above). The value rate of inflation rose sharply in the 1970s while the hours worked grew very slowly and indeed fell sharply in the 1974-5 and 1980-2 recessions. Thus the economy suffered what has been called ‘stagflation’.
The reaction of the monetary authorities to slowing value growth was to increase the money supply in order to boost economic activity, leading to an acceleration in the growth of the money in circulation. But this did not lead to faster growth in hours worked. As a result, the value rate of inflation was rising by more than 10% a year at the end of the 1970s (see Figure above). So the US monetary authorities, now under then Fed chair Paul Volcker, sharply changed tack and tightened monetary policy. The growth rate of money in circulation was more than halved by the end of the 1980s. Inflation (accelerating price rises) was replaced by disinflation (slowing price rises). Indeed, after the end of the Great Recession through the 2010s, inflation virtually disappeared (see Figure above).
Changes in value depend on the objective factors of a rising employed workforce (more hours) and a falling rate of profit (less growth in hours). But changes in money in circulation depend on the subjective reaction of the monetary authorities and the financial sector to these objective changes in the economy. The US Federal Reserve will ease its monetary policy if it considers the economy is weakening and will tighten its monetary policy if it considers inflation is accelerating.
In our view, it is the change in value creation, which is the objective factor in prices, that determines the subjective reaction of the monetary authorities. The monetary authorities believe that they can hold back the deterioration of the economy by increasing money supply in the wrong belief that more money can end any decline in investment and GDP growth. So the authorities increase the quantity of money faster than the quantity of value created. Thus with active monetary policy, inflation becomes a permanent feature of modern economies, although the rate of price inflation will vary, first because of changes in the growth of new value (hours worked) and second, because of the relative size of the reaction of the monetary authorities in varying the growth in the money in circulation.
Is there empirical support for the view that changes in value (hours worked) is the main driver of the value rate of inflation? First, there is a relatively high correlation between the rate of profit on capital and the change in hours worked (0.66). Second, there is a relatively high correlation between the change in hours worked and changes in money in circulation (0.60). Correlation does not prove causation. But these high correlations do indicate the ‘possibility’ of a causal relation. And there are strong theoretical reasons to assume that changes in value cause changes in money circulation rather than vice versa.
And we can add a statistical test of causation. A Granger causation test of the direction of causation between changes in value (hours worked) and changes in money in circulation finds that the ‘null hypothesis’ does not hold for changes in value leading to changes in money circulating, implying a causal connection, while it does hold for changes in money in cirdulation causing changes in value, implying no causal connection. This tends to confirm empirically the proposition that value growth is the objective driver of inflation and monetary injections are the subjective reaction of the authorities.[1]
To summarise, only our model of inflation is based on Marx’s value theory. We construct a value rate of inflation, as measured by the difference between the percentage change in money in circulation in the whole economy and the percentage change in value, namely the hours which have been expended for the production of commodities. Changes in value are the outcome of the interplay between two opposing forces: a rise in hours due to the expanded reproduction of capital (more workers, longer working day or year) and a fall in the growth of hours worked due to the increase in the organic composition of capital and the fall in the rate of profit.
So the value theory of inflation incorporates both the role of profitability and growth in investment as emphasised in part two of this series of posts by Mavroudeas et al, but also the role of money as raised by Shaikh in part two. Combined, we arrive at our theory. In modern economies, money is managed by the monetary system (central banks, commercial banks), with a measure of relative autonomy. The monetary authorities manipulate the money supply according to their assessment of the economic situation. This is the subjective element, which together with the objective element (the growth in value) determines inflation.
And here is the interesting implication. Both the official consumer price and GDP deflator indexes of inflation are highly correlated with our value rate of inflation. But the average annual value rate of inflation from 1949-2019 is much higher at 5.2% a year compared to 3.4% (CPI) and 3.1% (GDP def).
This suggests that workers’ real wages when measured in value terms (ie in hours worked to obtain commodities) have risen much less than when measured against official price inflation. We shall consider this issue in more detail in part four, which will also deal with why inflation spiked after the end of the pandemic slump in 2020 and whether inflation is now here to stay after the period of disinflation from 1982-2019.
[1] The results of the Granger causation test are available on request.
No comments:
Post a Comment