The Numeraire Jude Wanniski October 26, 2001

To: Website Students of SSU
From: Jude Wanniski
Re: The Numeraire

The lecture today is adapted from one I originally gave here in the fall semester of 1997. The idea that came to me then was the result of a back-and-forth e-mail discussion I had with Prof. Reuven Brenner of McGill University about the concept of money. As the discussion proceeded, we both realized that the term "unit of account" was not going to be sufficient to cover all we needed to cover. In a world of floating currencies, each currency is a unit of account within its own realm, but how do we tie together these variable standards of measure to serve the needs of world commerce. The term international economists have used is numeraire – a monetary common denominator.

In the 1944 Bretton Woods international monetary system that lasted until 1971, the dollar was defined as 1/35th of an ounce of gold, and each other currency was defined as some fraction or multiple of a dollar. The unit of account in each country was the national currency; the numeraire directly or indirectly common to all was gold. Because the dollar was fixed to gold, the dollar itself became the numeraire. When the dollar/gold link was broken, ending the Bretton Woods system, the official world of commerce continued to think of the dollar as the numeraire, but because the dollar now had no definite value, it was useless as a numeraire. It could not serve to tell us which currency was experiencing inflation and which deflation. Economists and the business and financial press would see other currencies "strengthening" or "weakening" relative to the dollar, but the more reliable numeraire remained gold -- the one monetary vehicle that enables us to see which currencies are inflating and which are deflating. If the dollar price of gold were falling and the Deutschemark price of gold falling faster, the press would report the dollar was "weakening," even though its purchasing power was increasing. In almost all important business reports on radio or television, there is mention of the hourly movements in the price of gold -- thus providing listeners with a meaningful benchmark.

In 1997, when we discussed the numeraire, the European Community was still a year away from its plunge into a common currency, a common unit of account -- the euro. But we have seen since that the euro is not as stable as it was expected to be -- against the dollar, which is the dominant accounting unit in the world. Why? Because the dollar, not the euro, is the currency in which gold and oil are priced everywhere -- precisely because the world of commerce needs a single benchmark so that the trillions of contracts being made during the course of a year can relate to a common reference point. In September 1987, Treasury Secretary James Baker III suggested to the International Monetary Fund that its member nations move toward a coordinated stabilization of their currencies -- using a "basket of commodities, including gold" as a "reference point." The suggestion was quickly forgotten when a month later the stock market crashed on Wall Street. The need for a "reference point" is obvious when you are talking about an international agreement of coordination -- which central bank should be tightening or loosening at any given moment in order to keep currencies from drifting apart in value? There must be a reference point. It would be a market signal, for example, that says because the dollar price of gold is rising, but the euro price of gold is steady, it is the U.S. Federal Reserve that is too "loose." At Polyconomics, all our country analysts follow the values of individual currencies relative to gold on a day-to-day basis, as a reliable guide to whether specific governments or central banks are managing their national accounting units in ways that reduce the risk to producers and transactors. This is the single most important advantage Polyconomics has had over all its competitors in the last 23 years, as we are alone in seeing gold as the constant and the paper currencies in constant fluctuation. It is the dollar that revolves around gold, not gold around the dollar.

It is my belief the term numeraire most properly applies to a unit of labor, the most basic economic unit -- not any mixture of labor and capital. In other words, how much time and energy does it take a man to do a unit of work -- digging a hole, for example. You cannot answer the question without specifying whether he has a spade or a backhoe, the latter being a mass of capital. In the world of floating currencies around us, the dollar is viewed as a unit that mixes labor and capital in amounts reflected in government statistics, which necessarily mix apples and oranges, spades and backhoes. Economists -- even Nobel Prize winners -- and financial journalists have been puzzling over why there has been no statistical inflation even though unemployment rates have been at 30-year lows. If they would have gold as their numeraire instead of the dollar, they would more easily see that the fall in the price of gold over the last five years -- to the $275 level from the $385 level -- has signaled a monetary deflation.

The idea that labor is the central measuring rod of value is certainly not original on my part, but is inferred from my reading of the classical economists, especially Karl Marx. The following is from Capital, quoted in Essential Works of Socialism, pp. 63-65, ed. by Irving Howe, (Holt, Rinehart and Winston, 1970):

The mystical character of commodities does not originate...in their use-value. Just as little does it proceed from the nature of the determining factors of value. For, in the first place, however varied the useful kinds of labour, or productive activities, may be, it is a physiological fact, that they are the functions of the human organism and that each such function, whatever may be its nature or form, is essentially the expenditure of human brain, nerves, muscles, and so on. Secondly, with regard to that which forms the groundwork for the quantitative determination of value, namely the duration of the expenditure, or the quantity of labor, it is quite clear that there is a palpable difference between its quantity and quality. In all states of society, the labour time that it costs to produce the means of subsistence must necessarily be an object of interest to mankind, though not of equal interest in different stages of development. And lastly, from the moment that men in any way work for one another, their labour assumes a social form.

Whence, then, arises the enigmatical character of the product of labour, so soon as it assumes the form of commodities? Clearly from this form itself. The equality of all sorts of human labour is expressed objectively by their products all being equal values; the measure of the expenditure of labour power by the duration of that expenditure, takes the form of the quantity of value of the products of labour; and finally, the mutual relations of the producers, within which the social character of their labour affirms itself, take the form of a social relation between the products....What, first of all, practically concerns producers when they make an exchange is the question, how much of some other product they get for their own? In what proportions the products are exchangeable? When these proportions have, by custom, attained a certain stability, they appear to result from the nature of the products, so that, for instance, one ton of iron and two ounces of gold appear as naturally to be of equal value as a pound of gold and a pound of iron in spite of their different physical and chemical qualities appear to be of equal weight. The character of having value, when once impressed upon products, obtains fixity only by reason of their acting and reacting upon each other as quantities of value.

These observations by Marx help explain his belief that Gold is “the commodity money par excellence.” When we know the value of gold money and agree upon it in relation to other values, we can then triangulate into other relative commodity values. If we know one ton of iron equals two ounces of gold, we can relate one ton of iron and 700 loaves of bread or 300 chickens and one ton of iron or two ounces of gold. Marx could observe that in the marketplace, everything for sale or trade was priced in fractions or multiples of gold ounces, even though no gold ever changed hands. Over the centuries, new goods never seen before came into the marketplace, including autos, radios, telephones, heart transplants, etc. Each in turn established its value relative to the numeraire. Even today, when no unit of account is officially defined as a specified weight of gold, the market uses gold as the numeraire to triangulate the relative value of real things, as opposed to paper money that floats without definition. And the numeraire remains identified not with capital, but with labor. In my discussion with Reuven Brenner, I made the following point:

Why is this argument important? I think because it nails down the importance of having a fixed unit of account. As long as gold is mixed up with capital and labor, it opens logical discussion to the idea that gold will hold back progress. If gold is simply tied to labor, then it cannot hold back progress. An ounce of gold today will buy 10 haircuts in Tokyo, 30 haircuts in New York, and 300 haircuts in Lagos or Calcutta. This is why the idea of PPP (purchasing power parity) never really works in practice. A man with a pair of scissors is like a man with a shovel looking for gold. The scissors and shovel are capital, it is true, but minimal considerations that can be ignored. In a country that is totally starved of capital, only labor exists, and in that country one man will pick up scissors to cut hair and the other will pick up the shovel and head for the hills. In the time it takes for a thousand barbers to cut 300,000 heads of hair, a thousand miners will produce a thousand ounces of gold, which can exchange for 300,000 haircuts.

It is important that gold be seen as this elemental unit of measure, at least in its role as the most monetary of all commodities. The several billion transactors on earth need a common starting point, from which to measure their value in terms of capital in combination with labor. You cannot measure two media with the same standard of measure -- you cannot measure a length and a liquid with the same ruler. The use of gold as numeraire is critical to people as they go about their daily lives. It is not enough for them to know the price of everything in dollars, the official unit of account, when that unit varies from day to day relative to gold, the unit of account that people have used for eons to calibrate relative values.

Gold has been the best proxy for all goods and services offered for exchange in the market universe for most of human history -- because it does best the job of serving as a unit of account. This function does not involve trust, except in the sense that the world marketplace trusts that gold will hold its value over time better than paper currencies or other competing commodities. At that level, it involves spot prices, which represent units of labor, not capital. If I make bread and you make wine, and one bottle exchanges for one loaf, they both can be priced in terms of gold, for purposes of making the exchange. It may take you only an hour to make a loaf and take me all day -- because you may be using a lot more capital than I am. Still, the basic loaf exchanges for a basic bottle and a basic weight of gold. Over millennia, new commodities are added into the galaxy, but instead of gold seeming to lose its edge as the best proxy for all, it becomes even better. Milton Friedman thought for a while that the dollar would be better, as managed according to his scientific principles, but he was absolutely wrong, as Prof. Robert Mundell predicted would be the case.

The issue is confused by those who argue gold has not maintained its purchasing power since the 16th century, because it could not buy a tv set back then, and now it can. An ounce of gold today can buy as many loaves of bread or bottles of wine as it could 2000 years ago, as long as we agree the bread and wine is made by labor, in the most basic way, as opposed to being made in high-tech bakeries and vineyards. Silver, once was competitive as a reliable monetary unit, but is no longer. For thousands of years silver and gold ran neck-and-neck, at around 15 ounces of silver to 1 of gold, but when silver was demonetized as "legal tender" in 1873 by both France and the United States -- joining England in a single unit of account -- it now takes 80 ounces of silver to buy the basic loaves and bottles that gold still buys.

This doesn't happen by magic. The golden constant is constant because the basic unit of labor remains constant. A man can dig the same size hole today that he could 2000 years ago, in the same amount of time. Because he cannot eat gold or use it as shelter or for clothing or transportation or watching the evening news, he only takes enough of it out of the ground to maintain its value as a unit of account. When he takes out one ounce too many, there is an incipient inflation. When he takes out an ounce too few, there is an incipient deflation. By tying our paper monetary system to this highly perfect gold market, we get the most efficient money -- for all its functions.

The most persistent question that arises about a return to a gold standard concerns the amount of gold available in a growing economy. If we make gold the money, is not economic growth going to be limited to the amount of gold we can dig out of the ground? With only several hundred million people on earth, maybe that was okay, but now with 6 billion and counting, are we not tying ourselves down to a glacial rate of growth? The answer is that we only use gold as the numeraire, because we cannot have an efficient world monetary system without one. A numeraire is like a yardstick, a unit of measure that is fixed in time and space, the length of a man’s stride. The amount of things you can measure is not limited by the number of yardsticks you can make. The amount of economic growth the world can have is not limited by the numeraire. Because every country in the world uses non-interest-bearing government debt as its money -- its unit of account, its medium of exchange, and its store of value, as long as it keeps the supply of debt that pays no interest equal to the demand for that "money" at a constant value to the numeraire, there can be rapid growth with no increase at all in gold stocks. We can measure a doghouse and a skyscraper with the same yardstick.

The arguments against a gold numeraire in the same way would insist that the world could be swamped with too much gold if the Soviet Union or South Africa should decide to mine gold like crazy and dump it on the world markets in exchange for dollars at a guaranteed fixed price. But if they supply one more ounce of gold than required to maintain the numeraire, the central banks would add liquidity to prevent the gold price from falling due to its excess supply. Follow what would happen: The Russians direct their gold miners to dig up another 10 million ounces of gold in a calendar year. We assume they are doing this because the United States is guaranteeing $350 per ounce, and the Russians want to get the $3.5 billion for the 10 million ounces. If they dug up 10 million ounces and sold them into the open market they might assume the gold price would fall and they would get much less than $3.5 billion. In the event, the dollar price of gold would be depressed to the bottom range of the "gold points." That is, a $350 fixing would mean we might buy gold at $345 and sell gold at $355. The actual operating mechanism would mean we would add liquidity when gold got to $345.01, to avoid having to actually buy the gold. The Russians would keep selling until all their gold was in private hands, having realized the lowest possible price.

The Russians now have $3.45 billion in dollars. What do they do with it? It is not likely they will buy U.S. Treasury bonds, because if they did, they would get the interest-bearing bonds and the U.S. Treasury would have the dollars. Instead of having to issue $3.45 billion in bonds to refinance the national debt, the Treasury has the cash. All that has happened is that Russian gold miners have worked like crazy, they have not been paid, and their government has U.S. government bonds. That’s why this scenario does not work. If the Russians have $3.45 billion in cash for their gold, they will want to spend it on things they can buy for dollars. Say they decide to buy wheat from our farmers. They give the farmers the money and they get the wheat, which means the dollar price of wheat climbs on the world market. The farmers get the money, and have the surplus dollar liquidity the Fed produced to prevent gold from breaking below $345. In this scenario, the best buy on the world market is gold at $345.01, which the farmers buy until it reaches $354.99. (The farmers may buy something other than gold, but eventually the surplus liquidity will come to buy gold.) At that point, the Fed will sell bonds into the open market until the surplus liquidity has been absorbed.

If you stand back and examine the transaction, the United States has gotten the best of the deal. It bought gold for $345 from the Russians and sold it for $355. Our central bank produced surplus liquidity and then extinguished it, with not the slightest tremor to the dollar. American farmers sold the equivalent of $3.45 billion in wheat for $3.55 billion in gold. The Russians have the wheat, but they must sell the wheat for rubles in order to pay the gold miners for mining the gold.

This scenario is precise, seamless. It is exactly why only noodlehead governments would even think of trying it. It is why the Amsterdam banks could maintain a gold standard in the 17th century, why the Bank of England could take over in the early 18th century, and why the United States could take it in 1913, at the outbreak of war in Europe. The only reason the United States left gold is because the Federal Reserve was creating liquidity in 1971 in an attempt to lower interest rates and expand the economy. They did so even as foreign central banks, which were collecting that surplus liquidity in Europe, were asking for gold in exchange. The Fed simply ignored the gold signal, on the advice of President Nixon’s demand-side economists -- both Keynesian and monetarist -- who believed gold was holding back economic growth, when the problem was the high tax rates.

You should now be able to see why the financial markets become so disturbed by the Federal Reserve’s attempt to contain an inflation which it assumes must be imminent because of the tight labor market. There can be no "inflation" or "deflation" with gold constant, although certain prices can rise and others fall because the nature of overall production in the economy is undergoing change. By raising interest rates in a mistaken belief he is nipping inflation in the bud, Fed Chairman Alan Greenspan is merely slowing economic growth in sectors of the economy that rely on debt financing rather than equity financing -- equity that rewards capital through higher dividends or capital gains, not interest rates.