So it goes

This is what I remember of Kurt Vonnegut: in Mother Night, the main protagonist is a Lord Haw-Haw type character, propagandising for the Nazis—although, all the time, he is really an undercover agent working for the democratic side. Later, he comes to wonder whether he hasn't been a better propagandist than a spy.

In Bluebeard, Rabo Karabekian, an abstract expressionist and escapee of the Armenian genocide, paints his greatest works using some cheap industrial paint that soon falls apart.

Cat's Cradle—the invention of a religion and a dictatorship, so that the eternally poor have something to live for and to fight against: ie so that they have a meaning.

In Slaughterhouse Five: Billy Pilgrim falling backwards through time and seeing the bomber plane on TV extract the shrapnel out of the dead, as if by magic, as it flies backwards over the battlefield.

Refresher course

A review of the WinEcon economics learning software, part one

With the eventual aim of undertaking some modest empirical descriptions of the economic transition in eastern Europe and the former Soviet Union over the past 15 years or so, concentrating—though not exclusively—on the topics of economic growth and the economics of labour markets, about a month ago, I took a week off work to study economics again from scratch.

The course I chose was the WinEcon software (version 7.1), which can only be bought and downloaded over the Internet. This process was quite straightforward. By right-clicking on the desktop shortcut icon so installed (a little red-blue-green bar chart), I was able to enlarge the display to fill the whole screen (640 x 460 screen resolution), making on-screen reading easier (mind you, this disables the program's calculator, I think). By right-clicking the “start” button in Windows desktop and then “properties”, it is also possible to prevent the Windows taskbar from intruding over the top of the WinEcon software.

During my study week, I completed about a quarter of an undergraduate economics course—that is, six out of 25 chapters: two on microeconomics (chapters 1 ands 2), two on macroeconomics (chapters 9 ands 10), and two on maths and statistics (chapters 22 ands 24). Later, I read chapters 11 (the circular income-expenditure economic model), 12 (theories of money supply and demand) and 13 (mostly Keynesian short-term macroeconomics).

Although quite standard in terms of content, chapters 1, 2 and 10 were reasonably thorough and lucid, and the bit-by-bit interactive presentational style helped to sustain interest.

Chapter 1 presents definitions of basic economic “oppositional” terms (macro vs micro, nominal vs real, positive vs normative, command vs the market) and develops a familiar definition of economics as the study of rational agents forced to choose between competing resource uses in conditions of scarcity. Some of the rather abstract ideas involved are usefully conveyed by means of concrete examples and game-like illustrations. The most interesting section, however, was the exposition of economic modelling, in which the process is broken down into the following stages:

• statement of the problem;
• whittling down of influences to leave us with a set of simplifying assumptions;
• development of a theory to answer the question(s) posed at the "problem" stage;
• testing the theory against evidence; and
• provisional acceptance of the theory, if it comes through the test.

These stages are then illustrated using two well-known economic theories: the Keynesian consumption function (household consumption is mainly influenced by changing levels of income) and the Fisher hypothesis (interest rates rise and fall with inflation).

What quantity of a good will buyers and sellers purchase and supply at each price? Chapter 2 familiarises the student with the basic tools of supply and demand analysis: how to construct supply and demand curves, and an account of the factors that affect each (for demand: product price, household income, prices of complements and substitutes, consumer tastes, the number of outlets and advertising; for supply: the product price, the price of inputs, the scale of taxes and subsidies, technology, the weather and the number of outlets); the difference between movements along the curves and shifts in the curves; the interaction of the two to produce equilibrium product prices and quantities, and how tools of this kind can be used to predict likely changes in price and quantity under changed market conditions ("comparative statics"; the examples used are polices for the support of domestic agricultural prices and other kinds of government intervention). Lastly, the concept of elasticity is introduced.

At this point, I switched to the macroeconomic sections of the program.

The second of the macro section, Chapter 10 runs through the main conceptual problems of measuring output in a national economy—seen as the level of productive activity overall, or as the total product of that activity. The first pitfall to avoid is to exclude transactions, such as transfer payments to pensioners or the unemployed, that do not represent payments for productive activity, as well as those that denote only changes in ownership of an item, since no new production is implied; the second is to count only new value added at each stage of the production process, from the extraction of raw materials to the sale of final goods—ie to avoid double counting.

The picture of the circular flow of income between households, firms, the government and external actors is built up gradually as the basis for the three alternative measures of national income accounting. (In this model, injections into the economy exactly match withdrawals from it, by definition.) The expenditure method measures the spending of households, government, investors and foreign buyers on goods and services in the domestic economy, to arrive at gross domestic product (GDP) at market prices. The output method measures what firms actually get for their production (and therefore have available to spend on factor services), and differs from the expenditure measure by subtracting indirect takes (those levied on the prices of goods and services themselves, rather than on factor incomes) and adding any production subsidies, to arrive at the gross value added (GVA) at basic prices. The income method aims to total up the payments received for the various factors of production (land, labour and capital); summing together the factor payments from firms to household, subtracting direct taxes on this income—which are siphoned off to the government—but then adding the redistribution to households of some of this tax via transfer payments, we arrive at the measure for personal disposable income (PDI). One of the advantages of having three measures of national income, each gauging monetary flows at different points in the cycle, is that they act as “checks” on one another, helping to reduce errors.

Next, the module details some of the practical problems of data collection and estimation for each of accounting method. Simplified layouts of the tables that are the end result of the data-collection process for each of the methods look something like this:

Expenditure	Output	Income
Consumption	Agriculture	Salaries, wages
households	Production	Firms' profits
non-profit institutions	utilities	non-financial
Government consumption	manufacturing	public firms
Fixed investment	Total production industry	private firms
Change in inventories	Construction	financial firms & other
Acquisitions less disposals	Service industries	Mixed income
Total domestic expenditure	distribution, hotels, repairs	GVA at factor cost
Plus exports	transport & communications	Net product taxes
Total final expenditure	business & finance	GDP at market prices
Minus imports	government & other	-
Statistical discrepancy	Total services	-
GDP at market prices	GVA at basic prices	-

Moving snapshot
Presented sequentially, these structural snapshots of the economy allow us to glimpse something of its changing character over time, to describe broad changes within it—for instance, many Western economies once dominated by manufacturing now predominantly specialise in services. From the information so compiled it is possible to get a picture of the structural features of the national income in the UK in the post-war period, as follows:

• nominal GDP had reached £1.7trn by 2004 (from £1.1trn in 1996);
• close to 60% of national income takes the form of wages (compensation to employees);
• exports make up a growing share of expenditure (19% in 2004)
• investment makes up a falling share (13% in 2004);
• aside from the government, business and distribution services are now the dominant sectors and their share of output is rising; and
• manufacturing, utilities and agriculture accounted between them for only a fifth of the UK productive activity in 2004 and the share of all of them was on a trend of long-term contraction or decline.

The reason that any of this is useful is that an accurate description of a country's economic structure, as well as the changes to its economic structure over time, is an essential basis for sound analysis, as well as for meaningful international comparisons.

Other useful topics in the chapter include instruction in two methods for calculating real output changes from nominal data (the first values output in later years at base-year prices, whereas the second multiplies the change in quantity of goods in later years, compared with the base year, by product weights established in the base year), as well as a brief look at alternative methods of national income calculation—gross national income (GNI; GDP plus net income from abroad) and net national income (NNI; GNI less depreciation)—and the possible grounds for the inadequacies of these measures for capturing accurately economic welfare more broadly.

That’s enough for one day.

72

If you want to work out how long it's going to take a country to double its income, you have to divide 72 by the country's average growth rate. Thus, at India's average annual growth rate of 1.8% in 1950-75, its economy would double in 72/1.8 = 40 years; China's, at 6% in 1975-2000, would double at a much foreshortened 72/6 = 12 years.

I also liked the one-size-fits-all coffee-cup lids as an example of the myriad kinds of rather bland, but cumulatively significant, kinds of innovation that characterise economic growth.

Movement without change

Neo-classical growth theory is an attempt to explain the conditions in which dynamically stable, or equilibrium growth, may be achieved. A key feature of the method of analysis used in neo-classical economic theory is the application of marginal techniques to the demand side (explaining commodity prices in terms of variations in marginal utility to consumers, for example) as well as the supply side (factor returns explained by their marginal products), rather than just to the supply side, as with classical theory.

Very broadly, the picture of the growth process envisioned in neo-classical theory is as follows. The level of savings in the current period is conditioned by the level of income from the previous period of production. This determines the size of funds available for current investment, all current savings being absorbed for this purpose. Investment may be used either to equip new workers with the same level of capital as all other workers, in this way maintaining the capital-labour ratio (capital widening), or to increase the level of capital per worker, in this way increasing the capital-output ratio (capital deepening). Capital widening can increase the absolute level of output produced, but not the rate of growth of output per worker. Capital deepening can increase productivity in the short run, but market mechanisms will adjust the relative prices of capital and labour in such as way as to encourage firms to economise of one or the other, bringing the capital-labour ratio to its long-run average—that is, the one consistent with dynamically stable, equilibrium growth—so that, without technological progress, only capital widening can occur.

In the long run, therefore, for a given level of technology, an economy will tend to grow at a rate determined by the growth rate of the population (the causes of which lay outside the field of enquiry of the model), because, assuming full employment of all factors of production, all other variables with potential to influence the level or rate of growth of economic output will adjust. To put this in another way, in the long run, only technological progress can permanently increase the rate of growth of output and income per head, because it increases the average product of labour for any given capital-labour ratio, raising also the capital-output ratio. Counter intuitively, the rate of saving has no effect on the long-run rate of growth of the economy.

To reach these conclusions, the neo-classicists make a number of simplifying assumptions; some of the most important ones are as follows.

1. Output is of a single homogeneous commodity. The level of output depends on the quantities of inputs of labour and capital, and is subject to constant returns to scale (ie by doubling the quantity of inputs of capital and labour, output is exactly doubled).
2. The supply of labour—which is also homogeneous in character—grows at a constant rate. This is the most important “exogenously determined” variable, the one to which the other variables, conditioning one-other within the confines of the model, must adjust. There is always full employment of labour, such that any extra contribution to productive capacity from this source translates exactly into actual contributions to output.
3. The rate of saving is constant. Whatever the level of income from the previous production period, in each new period, the same proportion of income is saved and the same proportion consumed. The current level of saving is determined by the level of income achieved in the previous period—that is, it is explained by processes at work within the model.
4. All current savings are employed as current investment in capital. This is so because interest rates on the capital market adjust to equate the two; if there is an excess of savings, interest rates will fall, raising the relative profitability and thus attractiveness of investment.
5. Capital does not to depreciate, but the “capital-deepening” process is subject to diminishing returns; that is, as the average capital per worker increases, the increase in output for each new unit of capital added to the production process is less than the last (ie the capital-output ratio begins to decline). The explanation for this for neo-classical economists lies in the low substitutability of labour and capital inputs.
6. The rates of reward to production factors are not constant; on the contrary, the movement of relative prices is the mechanism of adjustment by which a steady-state growth path is achieved.
   
   

Left well alone

Marx versus the nostalgists
Marx thought that inequality was the tool by which humanity would pull itself up by its bootstraps and in this way, eventually, escape the "realm of necessity"; only then would human history really begin—ie a history in which we would be able to break out of the stultifying "predestination" of the class system. But class inequality was the harness by which freedom would be earned.

Looking about, it is possible to spot potential new economic forms growing in embryo in the womb of the old society (eg open-source software development, where status among one's peers, rather than profit, seems to be the strongest motive for innovation). However, the old social relations are clearly still "forms of development" of productive power—which will have to be at an extremely high level if socialism is not to be maintained by force, if it is to be "stateless"—rather than its fetters.

Andrew Murray's kind of left, on the other hand—looking back for inspiration to the stark semi-poverty of the Soviet Union, with its bread queues and drab uniformity, the built-in "excess macroeconomic demand" or "sellers' market" that was the lot of your average Soviet citizen (in the good years, that is)—often seems to me to be one of the possible fetters, or brakes, on the broad trend towards progressive social change and worldwide material improvement.

Why so? How did this extraordinary situation come about? Is it not one of the most puzzling conundrums of the day?

Certainly, there are many things wrong with today's mainstream left, and Mr Murray's political praxis, I would argue, has as good a claim as anyone's to exemplify some of its least attractive features.

And yet I think that, despite itself, this nostalgic, slightly provincial, slightly chauvinist left may still serve a useful role—in as much as a thoroughgoing critique of its positions is likely to suggest what a useful left might look like, by way of contrast. Engels calls this "the power of the negative".

Instrumental anti-racism. Sometimes this school portrays free speech as merely a devious bourgeois device to protect the expression of racism. This seems to me an error of historic proportions, for when you give an inch to the state or to the dominant civic culture, it is more often than not the weaker social elements who get it in the neck, and to whom the restrictions or "protections" are thenceforth applied. Plus Mill's point applies: the more ideas, the more likely we are to come to good conclusions. Bad ideas had best be let out into the open, there to be hunted down. If only Mill had been able to put this dialectically, or, better still, in the unnecessarily complicated manner of modern French philosophy, I believe that it might have better caught on with this brand of radical conformism. In fact, far from opposing racism, of late, this left has teamed up with far-right racists, whose aim it is to impose their highly restrictive values on other Muslims.

Smash the machines! This strand of the left typically decries potentially poverty-reducing globalisation, instead of proposing an alternative kind of globalisation—one in which the undeniable benefits of international trade are more equally distributed. (By "exporting" good working conditions, for example.)

Down with this sort of thing! This leads me on to the chief failing of this clique: they have no plausible and appealing alternative to the "neo-liberalism" that they despise, but which they can't be bothered to understand—hence the disastrous faith-based appeals to "more of the same", the posthumous rehabilitations of the Soviet Union (and, by the way, the official Soviet policy of the equality of the races and sexes bore a similar relation to reality as the 1936 "Stalin" constitution to the everyday practice of Soviet democracy), and of the 1970s ("when unions was king"), the starry-eyed adulation of General Chávez, because "anything but this".

And yet, even with its unconscionable level of inequality, and the relatively harsh restraints it places on all-round human development—chiefly, perhaps, by the length of the working day—this society now, here in the West, is the best, economically and politically, the richest and the freest, that there has ever been in the 6,000 years since someone put up a clock tower in Uruk and said to themselves, "I think I might stick around"; which only shows what a long way there is to go and that we are probably living in an early stage of human history.

Two faces are better than one. Finally, the leftist of the kind I am thinking of tends to shed crocodile tears over the small successes of fascist parties at home, while carrying out extensive PR campaigns for similar or much worse ones abroad—by playing up the crimes of the imperialists, and playing down or skating over the crimes of the resistance death squads in Iraq, for instance. Hence my suspicion that anti-imperialism itself, at least in its current manifestation, is perhaps one of the most serious barriers to the development of some kind of humane and productive socialism in the future.

Frankly, I'd be very afraid of any kind of socialism in which this left had too much of a hand, as it has appallingly low standards for what a society based on the free association of the producers might be like, and it is usually quite prepared to trade off a bit more bread for a bit less freedom, whereas it seems likely that more of one leads to more of the other, and vice-versa.

Measure for measure

"for with the same measure that ye mete withal, it shall be measured to you again"

Economic growth is defined as the rate of increase of output over a set period of time. In the EU at least, following the widespread adoption of the European System of Accounts (ESA 95)—which the UK did in 1998, for example—the standard measure of economic development is gross domestic product (GDP) at constant market prices.

GDP measures the value of final goods and services produced in the domestic economy in a given period of time, usually one year; it includes indirect taxes, net of any product subsidies (this explains the “market prices” bit).

By stripping out from the money value of current production—or nominal GDP—any changes in the average price level in the economy between one year and the next, we are able to derive real or volume data. One method for doing this is to value the output of a later year using the prices prevalent in the earlier or base year. By comparing the change in the level of real GDP between periods, we are able to calculate real rate of change of GDP per year (this explains the “constant” bit). Other measures of output, less common nowadays, include gross national product (GNP), which takes into account the net flow of resources into or out of the economy, and national income (net national product, NNP), which includes depreciation.

Typical conceptual criticisms of GDP as a measure of economic growth are that it fails to take into account what is not bought or sold—valuable non-market services such as housework, for example. It also fails to encompass the benefits or costs associated with the growth process: the positive or negative economic utility value of leisure or pollution, for instance. Consequently, measures such as net economic welfare (NEW) have been proposed to overcome some of these problems, but have not tended to catch on—so far, at least. Additionally, any attempt to measure growth is compromised by the fact that the permanent innovation (which is intimately linked with economic growth) makes comparisons of the value of products across time difficult or, if a new product allows an activity that could not previously be undertaken, impossible.

Mr Ricardo's political arithmetic

The question of why economies grow has been central to economics since its emergence from the Enlightenment as a discipline in its own right. On this, I shall take an approach that will allow me to look at the question from different angles, bit by bit. The best place to start, it seems to me, is with a little foray into the history of economic thought.

All the classical economic theories were theories of growth—that is, they were designed to show that freedom for economic actors and free trade were essential for improving the prospects for the growth of national wealth. Later, with the "marginalist revolution" of the 1870s, mainstream economists shifted their focus to the analysis of the static problem of resource allocation. In the 1950s they began to apply marginalist tools to the problem of the growth of whole economies over the long term. More recently, this approach to growth theory has itself come under fire.

Of the classical growth theories, that of David Ricardo is perhaps the easiest to expound in a small space. As is usual in classical economics, Ricardo looks at the growth process through supply-side glasses to explain the development of productive capacity; this he assumes will be the same as actual output because market mechanisms assure full utilisation.

Ricardo's model is of a mainly agricultural economy in which production is the outcome of the combination of three factors: land, labour and capital. At any point in time, land is fixed in quantity but variable in quality; the labour supply is fixed; and the stock of circulating capital, or wage fund, is also fixed (it is envisioned as a stock of corn).

In any one period of production, the owners of capital hire the owners of labour to work on the land. Capitalists deploy labour in such a way as to equalise the marginal product of labour. The wage rate of workers is determined by the size of the wage fund, divided by the number of workers (the labour force); capitalists will hire workers as long as the marginal rate of labour is higher than the wage rate, the difference between the two accruing to the capitalist as profit; the income of landowners, in the form of rent, is determined by the difference between the average and the marginal products of labour multiplied by the number of workers on a farm.

This is the static part of the model. The model is set in motion by the behaviour of the economic groups with regard to their income. Workers consume all their wages to reproduce themselves, so that they are fit for labour in the next production period. Capitalists tend to save their profits, adding it to the volume of circulating capital, the wage fund—which is the source of economic growth. Landlords spend all of their income on "unproductive consumption".

However, in Ricardo's model—and in contrast with the model of Adam Smith—growth is not without inherent limits. This depends crucially on the role played by the expansion of the population, which is stimulated as a growing wage fund boosts the wage rate, raising it above subsistence level. As production on existing farms intensifies and also moves outwards to less fertile land, rents increase, which is useless from the perspective of economic growth; worse, the marginal product of labour falls and, with it, profitability, eventually bringing the growth of the wage fund—and thus the growth of national income—to a halt. This is known as the stationary state.

Although Ricardo’s model retains a certain self-contained elegance, it cannot be reconciled to the growth process as it has subsequently happened. This is for three main reasons.

• Population growth has often been "exogenously" determined, and the envisaged close relationship between the income and population growth has not obtained.
• There has been steady progress in the development of agricultural technology, some of it induced by the pressures of population growth.
• The emphasis on the role in growth played by circulating capital ignores the contribution to growth from the employment of fixed capital—particularly in industry, but even in agriculture—through improvements in productivity.

Under the old regime

Looking through some of my old stuff, I thought that a useful approach might be to "recycle" material from the past, updating it where necessary. Of course, some material does not need updating, since the situation it refers to has passed away. However, the definitions and tools may be of continuing use.

Repressed inflation in STEs
The interlinked phenomena of inflation and shortage in Soviet-type economies (STEs) can be attributed to a number of causes. Inflation is the movement of the aggregate price level for goods and services across the economy as a whole, although, when looking at causes, it is sometimes helpful to distinguish between those that are "one-off" in nature, originating from "outside" the normal workings of the economic system (for example, from mistakes of economic policy, or exogenous shocks, such as a rise in energy prices or a bad harvest), and those that are systemic, when the conditions behind the rise in the aggregate price level are constantly reproduced from within.

Causes of inflationary pressure under state socialism
In the classical model of state socialism, the system-specific causes of inflationary pressure may be listed as follows.

• Persistent labour shortages put upward pressure on wages which, if reflected in actual wage increases for the economy as a whole, will tend to increase production costs.
• As producer-sellers, firms have some vested interest in raising prices.
• Operation of a passive monetary policy may lead to some uncontrolled growth in the money supply.
• There is persistent excess demand at a macro level.

The last of these, excess macroeconomic demand, exerts the strongest inflationary pressure. And although there are also variations in the degree of excess demand prevalent in different sectors of the economy, it is most powerful in the state inter-firm sphere.

From this perspective, perhaps the most important phenomenon to understand is the process of taut or overfull employment planning. This is when the output plan for the enterprise is deliberately set above the firm's production capacity. The aim is to "seek out" resources; for individual managers, increased physical output is also the criteria for promotion. The effects, however, are that firms try to cut corners—for example, by reducing product quality or raising costs. From the point of view of structural constraints on the firm, this presents few problems, since the question of exit of firms from production is not an economic decision but a bureaucratic-political one. The peculiarity of the state property form means that, while in theory it is owned by everyone, in practice, "no one has a true inner interest in ensuring caution in handling money".

Soft budgets, passive money
In effect, the regime of the output plan means that a firm's budget is not a strict restriction on its access to resources. If a transaction seems to be demanded by the plan, a cheque—even one in excess of the firm's budget—will be cleared. Herein lies the "soft" character of financial budgets at enterprise level, and also the "passive" character of money in the inter-firm sphere. "Money" in this sense exists only as an accounting unit and does not exert any buying power. The total monetary demand for firms follows the financing needs of the plan. This is the reason why the term "excess demand" is sometimes considered inappropriate, and we should use the term remembering that it is compromised by the non-market character of the transactions involved. For this reason monetary and fiscal policy has no effect on the level of output. The rationale for Kornai’s observation that the firm under the classical socialist model has no particular interest in limiting its demand for inputs is thus seen to flow from the imposition of “taut” planning.

What symptoms when prices are controlled?
In STEs the symptoms of inflationary pressure are revealed by means other than price rises. This leads us to the question of repressed inflation, which may be introduced by a method of contrast. For while inflationary pressure may lead to a rise in the aggregate price level (open inflation), or actual rises in the prices of goods and services may not find their way into the calculation of the inflation index (hidden inflation), repressed inflation is said to exist when upward pressure on the price level is resisted by administrative control. (It is also defined by Nuti as the rate of change of excess demand.)

In the consumer goods market, too?
First, the boundaries of the debate must be delimited by saying that while the existence of excess demand in the state inter-firm sector of STEs is rarely challenged, the question of the endemic nature of repressed inflation in the market for consumer goods has been hotly debated. Also, the credit expansion typically exhibited in the state inter-firm sector would not have inflationary effects in the consumption sphere unless the (usually strict) separation of the functions of money between the two began to break down—if, for example, enterprise credit was allowed to be used to pay for wage increases and it became someone's income (ie if it was “activised”).

How, then, are we to identify inflation if movements in the price level are ruled out? To those who argue that repressed inflation is endemic to the consumer goods market, the evidence seems obvious: the chronic existence of queues, of shortages and of black-markets. These phenomena seem to testify to real attempts to consume more than is officially being produced. This is the kind of thinking adopted by Kornai in his attempt to construct an "index or partial indicators" from, for example, the number of building orders refused, unfulfilled car orders, or the length of waiting lists for accommodation. Pindak takes an equally direct approach to an analysis of Czechoslovakia between 1972 and 1978 by tracking the decreasing proportions of goods in the market for foodstuffs and industrial products not experiencing supply problems.

Another approach is to look at trends in saving. The theory here is that if consumers cannot make their desired purchases, and the bureaucracy prevents a rise in retail prices that would choke off demand, it is likely to be reflected in the rapid and involuntary growth of savings in relation to income or sales. However, since it is difficult to detect a change in the savings rate which signifies the qualitative change from voluntary to involuntary saving, some observers have pointed to the growth of the money-income ratio in many STEs as evidence of involuntary saving (and hence repressed inflation) which, over time, develops into a monetary overhang: frustrated buying power accumulates over the economy in the form of liquid assets (cash and sight deposits).

Charemza and Gronicki take another approach, using the rational expectations hypothesis to estimate excess demand for consumption and labour both in absolute terms and relative to quantities transacted, the results of which show a U-shaped pattern for excess consumption demand in Poland between 1960 and 1980, bottoming out in 1970, steadily increasing to the mid-1970s at about 8% of total quantity of consumption sales. They also employ a static indicator—the difference between the estimated market-balancing price and the real price—to suggest the strong growth in the level of repressed inflation in Poland after 1970.

From this perspective, macroeconomic rationing—and therefore repressed inflation—is seen as a quintessential and permanent feature of STEs, even in the market for consumer goods. First, because excess purchasing power is often deliberately built into wage settlements by the planners, thus forcing the adjustment process onto consumers; and second, because it is seen as a basis of social discipline (and political power).

The initial criticism of this position must be that visible signs of shortage cannot, without further qualification, be taken as indicators of excess demand at the macroeconomic level. This is because the root causes of the observed phenomena may be microeconomic in character. They might, for example, be the result of inadequacies of distribution rather than production, or they may stem from the rigidities of pricing policy (that is, a problem of relative rather than aggregate prices). These observable indicators may be particularly misleading if we think of repressed inflation as the rate of change of excess demand.

Why do we need to know?
The question of the source of repressed inflation is an important one, as it will influence the choice of stabilisation policies pursued in the period of transition from socialist to capitalist economic systems.

Is aggregation appropriate in STEs?
Within the field, there has been a somewhat heated dispute over the theoretical justification for statistical aggregation, given the nature of Soviet-type economic systems. An example of the aggregative method is an econometric study of Portes and Winter. While arguing that the roots of the problem are microeconomic in character, they maintain the pertinence of a macro approach to an analysis of economic imbalance. They test and reject the hypothesis that excess macroeconomic demand is endemic to the consumer goods markets of STEs. On the contrary, they conclude that from the mid-1950s to the mid-1970s excess supply was the dominant regime in three out of the four countries analysed (Hungary, Poland the GDR and Czechoslovakia).

Kornai, in contrast, contends that in a shortage economy, the idea of excess macroeconomic demand is not an operational category. By this, I think he means that when the consumer is forced to substitute for their original buying intentions, or is deterred from following through the buying intention in any form, the netting out of shortages for some products against surpluses for others is inappropriate, since withdrawal from the market reduces the level of observable shortages; also, perhaps, that quantitative measures fail to take into account qualitative criteria. This may mean that Portes's rhetorical criticism of estimates of excess aggregate labour supply in Western countries because of the phenomenon of the "discouraged worker" may be inadvertently justified.

Holzman's indicator for repressed inflation—the ratio of free market (kolkhoz) prices to state prices for foodstuffs, weighted by share of output and expressed as an index—assumes that the level of repressed inflation in the state sector has a direct impact on the level of prices in the free market. The ratio indicates a decline in the level of repressed inflation in the USSR for the decade after 1955, followed by a period of stability that lasted until the mid-1970s, when a definite, though undramatic, increase occurred until 1979. (However, I couldn’t see from the data what the proportions of the repressed inflation were at the base date.)

It has also been argued that trends in the money-income ratio should be interpreted as a wealth-income ratio in STEs, since the underdevelopment of capital markets means that only a narrow range of assets is available. The trends taken to indicate the development of involuntary saving may therefore be better explained by factors that interpret them as increased rates of voluntary saving. Ofer suggests that two such factors are the need to build up levels of saving: first, to be able to purchase consumer durables when no consumer credit is available; and, second, to offset the deterioration of public services and real levels of social security pay. In addition, a significant rise in the wealth-income ratio is hardly surprising given the very low levels of the ratio that prevailed in many STEs at the beginning of the 1960s. Cotarelli and Blejer suggest that, applying the life cycle hypothesis to consumption behaviour in the USSR, the increase in the wealth-income ratio could be interpreted as resulting from the deceleration of disposable income growth in the period 1965-80.

Conclusions
The hypothesis of endemic repressed inflation in the market for consumer goods thus emerges battered from the criticisms of more nuanced, more plausible interpretations of data and of rigorous econometric analysis. Nuti seems to add to the weight of this criticism by pointing out that the acceleration in the growth of liquid assets could be explained by the higher market-clearing prices on secondary markets, or by the necessity of speculative holdings in conditions of erratic supply—both of which would invalidate the concept of involuntary holdings for the sector as a whole. This might go some way, he suggests, to reconciling the estimates of low overall excess demand in the consumer sphere with concern for market imbalance. However, he goes on to argue that the bulk of the stored up buying power would be pressing on the lower priced, quantity-constrained markets as a result of (I think) the lack of substitutability between (luxury and everyday) goods. Consequently, the level of excess demand could be increasing fast in the state consumption sector, while open inflation in the non-state consumption sector keeps the level of excess demand for the sector overall at a stable level.