Subject: Re: Economist analysis
To: Jude Wanniski <firstname.lastname@example.org >
From Ben S. Bernanke, 8:40 am, 3/1/2005
Jude: Thanks for the helpful analysis. Ben
Jude Wanniski <email@example.com
To: Ben.S.Bernanke@* * * * *.GOV
Subject: Re: Economist analysis 02/28/2005 05:49 PM
Thanks for your response.
The reason there are "short-term deviations" in the 1980s is that the gold price is by far the best leading indicator of "monetary inflation," but the lead takes a long time to unfold in an economy where the maturity of the debt structure is quite long. If the average maturity of the debt structure, public and private, is 10 years, and the price of gold doubles next week, it takes 10 years for the CPI to catch up . Of course it doesn't take nearly that long for the bond market to price in the decline in the purchasing power over the near term. You can understand this best if you think of a hyperinflation, where the maturity of the public and private debt structure is measured in days, even hours. When the price of gold doubles in pesos or reals, the prices of goods and services doubles in days. This was the Brazil experience and the Argentine experience and, of course, the experience of Germany in the early 20s.
In late 1979, a year after I founded Polyconomics, the price of gold shot up dramatically, from $240 in the late spring, to more than $360, and I could see that Volcker did not know what he was doing. I lived in a modest home in Morristown, but called a real estate agent and bought an estate, 2 1/2 acres, with an Olympic swimming pool, in the most exclusive section of Morris County. I paid $185,000 and all my friends and family told me I would regret the purchase because I could not afford such an expensive place, including all its upkeep. I probably got the last 6% mortgage in Northern NJ, and within a year it was clear I could sell the place for $225,000. When I gave it to my first wife in our settlement, in 1986, she sold it for $550,000 and bought more modest digs. I tell you this in support of the infallability of my analytical model, which I developed on my own, building on what I had learned from Mundell. He's one of the great monetary theorists of all time, but he admitted to me he never thought of the variations in the maturity of debt structures in different currency markets.
It was a critical insight, which will do much to make you a believer once you let it roll around in your head for awhile. It did not come easily to me. One event helped me see it, involving Mexico. In September 1976, Mexico devalued the peso, from eight cents to five cents. I wrote an editorial for the WSJ, "The Nickel Peso," which warned that unless Mexico cut tax rates to offset the effects the inflation would have in rather quickly pushing the work force into higher brackets, they would find their budget deficit climbing and would conclude they had to devalue again... and again. They paid no attention, and in fact we got laughed at by Nobel Prize winners for running the editorial, but a year later Mexico announced it was devaluing again. It did so on a Friday afternoon. Monday's NYT had a story about how citizens swamped the shopping malls over the weekend, and found one man who had emptied his bank account and used it to buy several refrigerators at a department store. He told a reporter he was simply unloading pesos before they lost their value, as they had the previous year. The stores of course had a terrific weekend, selling everything like hotcakes, but when they called to order a replenishment of inventories, they found they had to pay twice as much for the fridges.
Your question, about how the gold price can give us a reliable signal in the short term, when it is only modestly above the CPI trend, is a wonderful question. Absolutely wonderful. The answer is that you don't need any more proof than the fact that as you have been raising the ff rate, the market has refused to shift up the yield curve. The market only sees a modest increase in prices in the short term and will not discount long-term price increases until they see the whites of their eyes. That is, the 13% increase in the gold price since 6/30/04 does not translate into a 13% increase in the 10-year yield because the market can't see what our government is going to do, what you are going to do, at the next FOMC meeting, let along 10 years out, so it does not discount the unknowable.
When I think of why you are so important in this analysis, Ben, it is because unlike Greenspan, you had no commitment to gold as a key signal. When he did not find gold reacting the way he expected it to, as he raised the ff rate six times in two years, he concluded that it was not a good signal and he could put it aside. I sent him a blizzard of memos back then telling him he had to forget the FF rate and shift to gold targeting, but at the time he was dizzy with the articles, editorials and books being written about him being the greatest Fed chairman in the history of the world!!
Bob Woodward wrote "The Maestro," a ridiculous book, as Woodward knows less than zero about money. Time magazine and others had Greenspan celebrated as the master of the universe, because he had defeated inflation -- just at the time I was warning Greenspan that he was presiding over a monetary deflation that would cause economic crisis and political crisis the world over.
The optimum way for you to go, Ben, would be to give a speech anywhere addressing the Economist piece, which is moronic as you say, and turning over in your mind various possibilities of what the heck is going on.... even acknowledging up front that you are puzzled and are looking for answers. Nobody will snicker at you, because Greenspan himself said so last week with his "conundrum."
Thanks for your memo. I told my wife it was the nicest thing that happened to me all day. As promised, I will not mention it to anyone else. I'll never embarrass you. My aim is to have you figure all this out and be the real maestro, who will be celebrated in story and song by future generations.
At 03:13 PM 2/28/2005, you wrote:
I agree that the Economist's analysis is moronic.
My concern about the CPI - gold link seen in your chart is that, even though there is clearly a long-term relationship, there are also wide short-term deviations as in the 1980s. So how can gold give us a reliable signal in the short run, eg, right now when the gold price is only modestly above the CPI trend? What am I missing?
From: Jude Wanniski firstname.lastname@example.org
To: Ben.Bernanke@* * * * *.gov
Subject: Economist analysis 02/28/2005 02:16 PM
You might have seen this from the Economist, but I thought you should not only see it, but realize how dreadful it is. (There is almost nothing I've seen in the Economist over the last 40 years that has to do with economics that has not been dreadful, but there is a big market out there for the neo-Keynesian paradigm.) The argument that real estate, share prices or asset prices in general should be part of an "inflation target" is the heart of the error. Prices rise because of prosperity or because of monetary inflation, and they fall for the same two reasons. Because real property in NYC or LA is rising is NOT NECESSARILY THE RESULT OF DOLLAR INFLATION. If the central bank believes it is, and it isn't, by raising interest rates to stamp out an incipient inflation it is instead stamping out prosperity. The Keynesian model does not draw distinctions between inflation and expansion or between deflation and contraction. Nor does monetarism. Note the Economist piece cites the "long pedigree" of its argument by citing Irving Fisher in 1911, the forerunner of Milton Friedman. It was Fisher who helped talked FDR into devaluing the dollar in 1933/34, which only caused a monetary inflation, tax bracket creep and a deepening of the Depression. Some pedigree!!
I'm happy, Ben, to hear from you that you are keeping an eye on the gold price and glad to get your thanks for the chart I sent of gold vs CPI since 1948. I'd be happier still if you asked a question or two about the chart.... so I could be assured you understand its significance and profound importance. You know I believe the Fed has been steering by a "faulty compass," but the Economist recommends an even more deeply flawed compass. The chart I sent should tell you that the monetary inflation (as separate from economic expansion) since 1948 had been totally absorbed by the general price level as of June 30 last year. You then began raising the funds rate at a measured pace to intercept an inflation that had been deadened at the very time you decided to kill it. The Fed's faulty compass since then has helped reintroduce a brand new inflation alongside the prosperity that has been encouraged by other economic policies of the government.
Please think about this. Look at the chart again and try to grasp what it is I'm saying. I'm having other charts developed with other inflation indices so you will see this not a flawed method. It would be helpful to the global economy if you presented some of your associates at the Fed, staff economists in particular, and asked them to mull over this material and find fault with it.
* * * *
STEERING BY A FAULTY COMPASS
Feb 24th 2005 Economist
Are central banks watching the wrong measure of inflation?
LAST week, for the first time, America's Federal Reserve published its forecast of inflation over the next two years. Many observers took this as a sign that the Fed had moved closer to setting an inflation target, as many other central banks have done. The minutes of the Fed's February meeting, published this week, confirmed that its policymakers have discussed the idea. Advocates of targeting argue that it would increase the transparency of America's monetary policy and maintain its credibility after Alan Greenspan retires as Fed chairman in 11 months' time. But at which measure of inflation should the Fed take aim?
It is widely agreed that central banks' prime goal should be price stability. Judged by existing consumer-price indices, central banks have indeed tamed inflation. But there are important things that such indices ignore--homes and shares, for instance. And the prices of these have been rising fast in many countries. Central banks should worry about those as well.
When inflation targets were first introduced (in New Zealand in 1989), the exact measure of inflation did not matter much. The main objective then was to reduce high rates of inflation by anchoring expectations. Today, however, consumer-price indices are arguably too narrow. Charles Goodhart, a former member of the Bank of England's Monetary Policy Committee, has long argued that central banks should instead track a broader price index which includes the prices of assets, such as houses and equities. America's core rate of consumer-price inflation (excluding the volatile prices of energy and food) rose by 2.3% in the year to January. But house prices rose by much more, 13%, in the year to the third quarter of 2004 (the latest official figures available).
These are excluded from America's consumer-price index (CPI); instead the cost of home ownership is represented by rents. But this can be misleading: over the past year, rents have risen by just over 2%, a lot less than house prices.
Ian Morris, an economist at HSBC, has devised a broader index, which includes house prices, giving them a weight of 30%, the same now attributed to the notional "rent" paid by homeowners in the core CPI. In the year to the third quarter, inflation as measured by this broad index was 4.9%, more than twice the rise in the core CPI. Assuming that house-price inflation has continued at the same pace, THE ECONOMIST calculates that broad inflation is now 5.5%, the highest since 1982 (see chart). Moreover, during the past three decades the gap between the two measures of inflation has never been so wide for so long.
If inflation in America is really higher than the official index suggests, then interest rates should also be higher. Similar measures would show inflation well above the standard figures in many other countries too. The main exceptions are Japan and Germany, where house prices have been falling.
A LONG ECONOMIC PEDIGREE
The idea that central banks should track asset prices is hardly new. In 1911 Irving Fisher, an American economist, argued in a book, "The Purchasing Power of Money", that policymakers should stabilise a broad price index which included shares, bonds and property as well as goods and services. Central banks already take account of asset prices by estimating their effect on wealth and hence on demand and future inflation, but the idea behind a broad price index goes much further, acknowledging that asset-price inflation can be harmful in its own right.
The most obvious way is through a giddying rise and subsequent crash of markets for shares or property. Big swings in asset prices can also lead to a misallocation of resources and so slower economic growth, just as high rates of general price inflation distort economies by blurring relative price signals. For instance, soaring property prices can encourage households to borrow too much and save too little, and can pull excessive resources into property at the expense of other forms of investment.
More fundamentally, if inflation is defined as "changes in the value of money", then the consumer-price index is flawed because it only measures the prices of current consumption of goods and services. A classic paper written in 1973 by two American economists, Armen Alchian and Benjamin Klein, argued that people care about changes in the prices not only of the goods and services they consume today, but also of what they use tomorrow. Because assets are claims on future goods and services, their prices are proxies for the prices of future consumption. If I buy a house--ie, a claim on future housing services--and its price is higher than a year ago, then surely that should be included in inflation since it reduces the purchasing power of my money. Many consumer durables, such as cars, which also provide services over several years, are already included in the CPI.
If the prices of goods and services and those of assets move in step, then excluding the latter does not matter. But if the two types of inflation diverge, as now, a narrow price index could send central bankers astray. Granted, asset prices are hard to measure: a rise in house prices may partly reflect an increase in the average quality of homes; and economists disagree over what weight house prices should have in a broad index. Yet buying a home is an enormous expense, so it is absurd to use such a rough approximation as rent, as does America's CPI, or to exclude the costs of owner-occupier housing altogether, as does the European Union's harmonised index of consumer prices.
Given the elusiveness of a perfect price index, central banks should keep using conventional, narrow inflation targets, but be prepared to undershoot them temporarily if house or share prices soar. This means taking a longer view than hitherto: an inflation target looking only two years ahead is too short-term. And although the calculation would be tricky, central banks might usefully publish broad price indices, including asset prices, beside existing measures. They would then find it easier to explain why they must sometimes raise interest rates when conventional inflation is low.
See this article with graphics and related items at