Milton Friedman famously claimed, “Inflation is always and everywhere a monetary phenomenon.” Does this relationship also hold in reverse?
Decades ago, when everything and everyone from unions to cartels was blamed for inflation, Friedman rejected the conventional wisdom and posited on the basis of empirical data that money supply drives price levels. He argued that prices increased not due to price and wage increases, but because the federal government made the supply of money grow faster than the real economy created value. This groundbreaking theory, while highly controversial and almost revolu- tionary at the time, appeared to be vindicated by the “Great Inflation” of the 1970’s, and has since become the core tenet ofmonetarism and modern policymaking. However, in a mark-to-market world, price may act insidiously to drive money supply and amplify boom-bust cycles.
Despite the copious amounts of money printed by the U.S. government through the fall of 2008, asset prices continued to fall precipitously. Relying on the assurance that Ben Bernanke would avert deflation by printing money, as indicated by Friedman’s theory, investors betting on hyperinflation were caught leaning the wrong way. Few recognized that credit contraction caused by price declines would annihilate money in a mark-to-market world. The contraction of total money supply overpowered the printing presses of monetarism, and cataclysmic deflation ensued. The Fed has long downplayed the role of asset prices in monetary policy. Yet it is apparent that in a credit-based economy that appraises assets on a mark-to-market basis, asset price inflation creates money and asset price deflation destroys money.
Imagine a marginal transaction that raises the price of an asset — say a house sells in San Francisco for 10% more than it sold for a year earlier. All similar homes in that neighborhood get marked up. Credit institutions then willingly lend against these houses at their new market value – their mark-to-market prices. In this way, small increases in price create vast amounts of collateral, which in turn be- get credit, liquidity, money velocity, and eventually total money. The net effect is that the appreciation of asset prices leads to an expansion of the total money supply.
As price increases lead to an increase in the amount of money available to bid on assets, such as our house in San Francisco, these perverse incentives promote further inflation in a “feed-forward” manner: anticipation of future price increases prompts higher bidding. The irony here is that as assets appreciate in price, they actually become more of a bargain, since these assets become scarcer relative to the money supply available to purchase them. This secondary effect is purely monetary and independent of the feed-forward effect of expectations regarding inflation. The potential explosiveness of the vicious cycle of per unit inflation and increase in total money supply is mitigated by human innovation that renders scarce assets more abundant through production i.e., more houses get built.
Conversely, as asset prices decline, the mark-to-market basis of the credit valuation precipitates a dramatic reduction of collateral, leading to a contraction of credit, liquidity, money velocity, and eventually total money. In our example, as houses sell for less all homes are assumed to be declining in value, banks are less willing to lend, and markets eventually freeze up as money is no longer available for buying homes. Price declines and consequent contraction of money creates a feedback loop. No matter how inexpensive they get, homes sold today aren’t bargains if they’re going to be cheaper tomorrow. Although the price tag of an asset might be lower, the decreased availability of money for bidding would also cause assets to become more expensive relative to the money used to buy them. Counter- intuitively, as assets fall in price, they may become less of a bargain.
This phenomenon may help explain the seeming contradiction in purchasing behavior that people have pointed out during this recent deflation. The world seems to be on a half-off sale, yet few parties are behaving as if the deal is a bargain. Asset investors note that assets are falling in price, yet lament the paucity of money to support bids.
When multiple asset classes deflate simultaneously, the feed-forward effect of price declines on total money supply can be dramatic. When asset prices in emerging markets and U.S. equity markets joined the housing markets in decline, American policymakers followed Friedman’s script and immediately began to increase money supply to combat the specter of deflation. Many investors similarly weaned on monetarist theory reflexively shorted the dollar and took long positions on commodities.
As these trends gained momentum, inflation lurked during the first half of 2008 due to rocketing commodity input prices. Notably, faith in monetarist policy amongst investors actualized the monetarist credo that an increase in money supply would cause inflation…for a while. Alas, commodity prices peaked in summer of 2008 and soon joined other asset classes in decline, and through the end of 2008 and the beginning of 2009, asset prices continued to fall precipitously in spite of the continued printing of money by the U.S. government.
A mark-to-cost model for asset appraisal, such as that seen for capital gains tax treatment, would substantially mitigate the insidious feed-forward effect of asset price movements on total money supply. In a cost-based appraisal system, only the house actually sold would be marked up in value; the other houses in the neighborhood would continue to be valued at their purchase prices until sold, and no money would be loaned against their “market value.”However, since a mark-to-cost model would be difficult to implement — it would not accurately reflect long term trends, such as a house in San Francisco owned for over a hundred years with a cost way below the market average — a more moderate solution such as asset valuations based on historical trend lines may be more practical. Under this scenario, banks would use an appraisal rate based upon the historical appreciation of homes in the neighborhood, over some set number of years to value the home for lending purposes, rather than the market value at any given moment in time based only on the most recent sales.
This change would seem problematic for America — we are a debtor nation, both to ourselves and to other countries. Stabilizing total money supply at low levels of money velocity could leave the country with insufficient total money to pay off our debts. It would seem that the U.S. remains on an implicit path to print enough money to allow us to inflate our way out of the current crisis, and at some point this policy will create the illusion of success. Eventually, however, the issues discussed above will once again resurface. Price increases will beget the whole cycle of money creation again, initiating the next boom-bust cycle.
The risks involved in implementing a new model for pricing assets may be high, but the risk of ignoring the issue may be a lot higher.
Originally published at The Journal of the Palo Alto Institute in Summer 2009.