Appendix
Monetarism: Never Mind The Quality
“It is not easy, it seems, for men to apprehend that their money is a mere intermediary, without significance in itself” – John Maynard Keynes [476]Section 4.5 mentioned that monetarists believe an economy can be regulated by appropriate adjustment of the money supply. Section 5.4 detailed some of the difficulties of applying this approach to a real economy, but a fuller explanation of the subject follows.
In 1911, USA economist Irving Fisher first proposed that an economy can be regulated via its money supply when he published his so-called ‘quantity equation’: MV = PT. Nice and simple, not unlike an alphabet block arrangement, but (as will be shown) requiring very sophisticated assumptions. In the words of the eminent monetarist Milton Friedman, M stands for “the amount of money in existence… [V,] the average number of times per year (or other unit of time) that each dollar [or other unit of currency] is used in effecting a transaction – the transaction velocity of circulation… [P,] an average price… [T,] an aggregate quantity of goods”[477] or transactions (volume of trade).
The quantity equation can best be demonstrated using a hypothetical example. Consider a tiny economy of two people, who maintain their economic flow with only two dollar notes each (M=4). Assume they engage in a total of (T=) 100 transactions over a year, each involving a transfer of an average of one dollar each (P=1); this means that each dollar note is used an average of 25 times over the year (V=25). The equation becomes MV = 4x25 = PT = 1x100.
In Fisher’s view, velocity (V) remains fairly stable, and hence the quantity equation lends itself to practical use: if the volume of trade (T) stays constant but prices rise (or fall), then according to the equation, the money supply must be growing (or shrinking); to stop prices rising (or falling), end the growth (or shrinkage) of the money supply. In other words, stable prices can be ensured, for a constant volume of trade, by a fixed money supply (or for an increasing volume of trade, by a money supply growing at the same rate).
Fisher’s views were widely accepted by economists until Keynes disputed them: he claimed that changes to the money supply affect the velocity of money (V) more than prices. (…palm grease…) “In the extreme form of this approach – as much a caricature of Keynes’s thinking as the attribution of a constant V is of his predecessors’ – the quantity of money supply did not matter at all for the course of prices or real output but only for the movement of V.”[478] Keynes’ views nevertheless held sway until the 1950s, when Milton Friedman began to revive Fisher’s ideas and develop them into what has since been called ‘monetarism’.
Friedman made his most influential case for monetarism in 1963, when he and Anna Schwartz published a lengthy study of price and money supply variations occurring in the USA over almost a century.[479] They interpreted their vast statistics as indicating that the velocity of money is not as flexible as Keynes had thought, and that Fisher had basically got it right. In Friedman’s view, V, the “velocity of circulation, though not constant, is fairly predictable”.[480] Hence, for stable prices, he advised, like Fisher, that the supply of money be increased at the same rate as the volume of trade; if prices rise, then money growth should be slowed below trade growth; and if prices fall, money growth should be increased. Friedman even offered an explanation as to why these variations to the money supply should have such effects.
Money growing faster in amount (M) than uses (T) has limited practical value; hence, people tend to trade it more freely to gain something ‘better’, maybe even being willing to offer or accept higher prices. Monetarists believe that more money growth than trade growth does prompt people to offer and accept higher prices, at least for assets like “bonds, equities, houses, and other physical capital… such… as durable consumer goods, and other real property”.[481] Essentially, according to monetarism, L$D causes the ‘price’ of oversupplied money, in terms of assets, to fall – meaning prices of assets rise. For the same reasons, interest rates tend to drop.
But higher-priced assets and lower interest rates “in turn encourage spending to produce new assets.”[482] Hence, with greater spending and more work, eventually, income rises. According to Friedman: “On the average, a change in the rate of monetary growth produces a change in the rate of growth of nominal income six to nine months later.”[483] Higher income, in turn, encourages more consumption, meaning more transactions (greater volume of trade), and this, in turn, entices greater output. But higher income, consumption and output correspond to raised demand, which increases prices in general, not just those of assets – although it takes about a year, from the time income increases, to have its effect. Thus, according to monetarists, after a time lag of a year and a half or more, ‘excess’ money produces inflation.
Likewise in reverse, though it takes longer to reduce than boost prices. In Friedman’s words: “If the rate of monetary growth is reduced, then about six to nine months later the rate of growth of nominal income and also of physical output will decline, but the rate of price rise will be affected very little. There will be downward pressure on prices only as a gap emerges between actual and potential output”.[484] In other words, at first, because of undersupplied money, production lowers, meaning unemployment and less trade – in Friedman’s view, on the evidence of history, unavoidable “painful side effects”.[485] However, at some point, lowering production any further becomes less of an option than slowing down price rises. Inflation eases – or as Friedman put it, the economy is cleared of “price system… static”[486] – “on the average about a year after the effect on nominal income and output, so that the total delay between a change in monetary growth and a change in the rate of inflation averages roughly two years.”[487]
Friedman based his claims and explanations on his analysis of data published in his book with Schwartz, and elsewhere. But not everyone is convinced. Harry Johnson claimed that Friedman’s analysis “involves a very dubious attempt to reconcile the hypothesis with the facts by some intricate inferences about the lags involved”.[488] Balogh called it “a (not-so-very-good) empirical correlation between the volume of money and prices” involving “curious statistical manipulation”.[489] Balogh also pointed out that even a superb correlation could “by no means prove the causal connection. Tuberculosis sanatoria were numerous when tuberculosis was rampant. Their number declined when tuberculosis as a general menace was eliminated. But it was not the suppression of sanatoria which caused the malady to disappear.”[490]
Friedman’s ‘evidence’ seems to me, as to many, inconclusive as regards cause and effect. For instance, in several graphs[491] showing growth rates of prices (consumer price index) and money supply, for various countries between 1960 and 1978, increased rates of growth of money in some periods were not followed about a year and a half later by increased inflation, as the theory suggests, but rather by reduced or unaffected rates. On occasions, decreased money growth also defied theory by being followed by increased or unaffected inflation rates. While Friedman admits that the timing can vary, on his evidence, the opposite of the proposed relationship seems just as plausible: that price changes alter the money supply. (…tin…)
Friedman, though, provided what he calls a “dramatic” example to support his belief. During the American Civil War, inflation in the South ran at ten percent per month from October 1861 to March 1864. “In May, 1864,… the stock of money was reduced. Dramatically, the general price index dropped.”[492] Not two years later, as Friedman’s theory predicts, but in weeks. Of course, much that happens in war cannot be expected to recur in peacetime. Nevertheless, Friedman’s views receive less contestable support from more recent events.
To pay for its military adventures in Vietnam, the USA resorted to money-printing, which boosted the growth rate of the money supply. But because the USA economy had reached full capacity by the early 1960s, production still could not increase enough to supply all of the military’s extra requirements – so, the military competed with the rest of the economy for whatever they could get. Thus subject to raised demand, the prices of physical resources needed to build bombs, tanks, uniforms, body bags, coffins, and all the other paraphernalia of war, rose – as did the wages of those involved – and also, after a time lag, inflation, “from 2.2 percent in 1965 to 4.5 percent in 1968.”[493]
Given this example, Friedman’s explanation of how money supply changes lead to altered prices does seem plausible – yet still an equal claim can be made for the opposite process. Certainly, to avoid impossible rates of turnover, the money supply, to some extent, must follow prices. But probably, the money supply and prices, and the volume of trade, follow each other simultaneously – to differing degrees depending on time and place – and balance via the undeterminable V.
That reality at least does not have to follow the simple L$D-inspired monetarist viewpoint can be demonstrated by a hypothetical economy having stable prices and money growth that matches trade growth. If the money supply suddenly grows a bit quicker than trade, according to Friedman, this “raises the amount of cash people (or businesses) have relative to other assets… [and t]he holders of the excess cash will try to correct this imbalance by buying other assets.”[494] Of course, some people may do this and settle for higher prices in the process. However, others may just use their extra cash to repay debts, or to increase their savings, or to spend in circumspect ways that do not raise demand and thus do not affect prices, but rather slow money’s velocity. Certainly, central bank policies since the Great Recession have boosted money supply significantly yet with demand and inflation barely changing.
How people actually behave depends on their expectations and situations, and on the general economic climate and mood of the day. Even if most spend ‘excess’ money primarily on consumption (as happened with the USA’s money-printing of the 1960s), prices (probably) rise only if the economy is operating at or near full capacity; otherwise, greater trade and/or lower money turnover without inflation (more probably) result. The quantity equation even allows for the possibility that an ‘excess’ of money could increase trade and/or decrease turnover enough for prices to actually fall.
Certainly, in suitable cultural and economic climates, what monetarists would regard as ‘excess’ money has circulated without causing inflation. According to Galbraith, for instance, in the USA’s ‘frontier’ days, especially “in the early decades of independence,… settlers… were… enthusiastically devoted to the creation of banks and by them of money… ‘[C]orporations and tradesmen issued ‘currency’. Even barbers and bartenders competed with the banks in this respect… Nearly every citizen regarded it his constitutional right to issue money.’”[495]
For the majority of the time from a decade after Independence until shortly before the Civil War, USA prices fell. To achieve such a result, monetarism requires, with a constant velocity of money, a volume of trade that grew, for most of the eighty-odd years, a little more quickly than did the money supply; while possible, the largely unregulated creation of money that did go on seems unlikely to have worked so consistently. More likely, with plenty of work to be done, with resources freely available, and with an economic capacity not even determined let alone reached, frontier expansion coped with frequently ‘excess’ money by boosting trade (and/or slowing turnover) rather than by increasing prices. Our current developed economies perhaps work less flexibly, but they are not, by any means, totally inflexible.
As for reducing inflation by decreasing the growth of the money supply, this also cannot be guaranteed – people’s overall spending habits may be unaffected, maintained by expectation of profits and/or by savings. Monoligopolies, in particular, can resist lowering or stabilising prices, even in opposition to general trends. A sharp enough cut-back of money (especially via interest rate increases) will eventually stop inflation, but only after the side effects of unemployment and reduced trade begin to dominate – somewhat like curing a sore toe by amputation at the knee.
Ultimately, the furious turbulence of the economic flow, with its various alliances and its misleading L$D-prices, has too complex and dynamic a set of interactions to control by simply topping it up with more money or draining some off. Because people make the economic decisions, prices depend on much more than merely the size of the money supply. As just one example, some of the changes to inflation rates before, during and after the 1991 Gulf War had nothing to do with the availability of money at the time – or two years earlier – but everything to do with largely unwarranted panic and fervour gripping sellers and buyers of oil.
The volume of trade or output also does not depend primarily on the money supply. Even in Friedman’s view: “Monetary changes affect output only in the short run – though ‘short run’ may mean five to 10 years… What happens to the output in the long run depends on such ‘real’ factors as the enterprise, ingenuity, and industry of the people; the extent of thrift; the structure of industry and government; the relations between nations; and so on.”[496] Most probably, the same ‘real’ factors, as well as consensual and personal tunnel-realities, have a greater effect than does the money supply not just on output but also on prices – in the short and long runs. So, at best, monetary regulation can fool only some of the people some of the time.
However, even if the monetarist formula could be treated as appropriate – which means that people’s complex pricing behaviour would be regarded as capable of being altered predictably by simple money supply changes – manipulating a real economy in accordance with the quantity equation cannot easily or accurately be done, because the equation’s variables have no easily determinable real counterparts.
Consider M: how much money does a nation have? Cash and demand deposits must be counted – in total called M1. If we also include term deposits, the total is called M2. But whether term deposits form part of a circulating money supply depends on whether they are treated as locked away in banks for months and years, or loaned to borrowers via fractional reserve credit. Foreign currency and building society deposits complicate the matter further – should they be counted as money? What about credit card accounts, commercial bills (IOUs), stocks and bonds, derivatives, vouchers and discount coupons? “One monetarist listed no fewer than seven… definitions of money… [each differing by the] volume of short-term assets included… [which] do not have much to do with transactions except when they themselves change hands”.[497]
Once agreement is reached that a horse has no udders or horns or fins, the beast can be attempted to be tamed, but even if a definition of money could be settled upon, accurately altering its growth involves certain practical difficulties. Monetarists advise that the central bank sell or buy bonds, and/or banks make more or less credit available. Neither process ensures that people change their borrowing, spending or pricing habits. Altering interest rates has a greater chance of enticing changes to behaviour, but while most of the efforts of governments to control the money supply since the 1970s have indeed concentrated on variation of interest rates, monetarists warn against it. In Friedman’s words, “more rapid monetary growth at first tends to lower interest rates. But later on, as it raises spending and stimulates price inflation, it also produces a rise in the demand for loans that will tend to raise interest rates… [And conversely, also.] This inconsistent relation between the quantity of money and interest rates explains why monetarists insist that interest rates are not a good guide to monetary policy.”[498]
But not only M causes trouble. T, too, has its problems. In a large complex national economy, “T, the total transactions, is all but unknown. Thus, a proxy, the national income or product [GDP], has to be substituted for it… [But this makes] the velocity (V)… an illicit hybrid concept”.[499] In practice, GDP is measured as the “value” (price) of the total “quantity of final goods and services produced during a year”,[500] and does not include the “value” of any “non-income transactions (house purchases, other transfers of assets, realization of paper profits etc.)”.[501] But because both sides of the quantity equation must deal with the same thing, to use GDP as T, V must become VGDP, “the average number of times that the money stock is used for making income transactions (that is, payments for final goods and services).”[502] However, VGDP will be affected by non-GDP activity, by changes to the velocity of money used in non-income transactions (whichever of money’s seven definitions is adopted). For instance, generally higher turnover results from booming (non-GDP) stock and property markets; whereas booming black markets reduce VGDP. So V – the only one that can be used – is not constant, nor predictable. Furthermore, using GDP as T makes a P of retail or wholesale price indices inappropriate because they include non-income transactions such as for property.[503]
Even if an economy had a static V and a measurable and controllable M and P, and people did respond to changes to M consistently (which asks a lot), to slow the growth of the money supply to less than that of the economy first involves knowing how fast the economy is growing. Formal figures for economic growth are usually produced quarterly, and take a few weeks to assemble and calculate (and are then usually adjusted the following quarter). So, guesses aside, any changes to the money supply cannot be made until about four months or more after known changes to economic growth; and they will be based on average quarterly figures that may hide significant daily, weekly or monthly fluctuations. Very easy to fall out of synch with such an arrangement, especially if two-year time lags apply as monetarists claim.
So, MV may equal PT, but the claim that this can be utilised to economic advantage seems MT. Supply-sider George Gilder neatly summed up monetarism’s weaknesses when he wrote that “monetarists cannot conclusively answer the question of whether monetary discipline is always possible, or even desirable; or whether the price level can be easily measured by the conventional tools; or whether the money supply can be easily defined or controlled… A perfectly valid theory may be irrelevant if the real world fails to offer the clear signals and instruments needed to apply it, or if the factors which it treats are not the controlling ones.”[504]
Yet occasionally, monetarism can work as expected: if enough people strongly believe that altering the money supply will affect prices, it can encourage them to act in ways which confirm their beliefs – for a while at least. Friedman recognised this enough to recommend that, to deal with the ‘side effects’ of the monetarist ‘cure’ for inflation, the “most important device… is to slow inflation gradually but steadily by a policy announced in advance and adhered to so it becomes credible… [This] is to give people time to readjust their arrangements – and to induce them to do so… [by changing their] anticipations about the likely rate of inflation.”[505]
So, the practical effectiveness of monetarism, like money itself, depends greatly – probably most greatly – on what is believed. Somewhat undermining its theoretical capacities, however, belief in monetarism itself has waned considerably since its heyday of the 1970s and 1980s, although governments and central banks, despite Friedman’s advice, still vary interest rates in their attempts to deal with inflation and unemployment (see section 5.4).
 Chapter 9 |