Commentary

Is the Fed Obsolete?

By Bob Gelfond
This article was prepared for the Cato Institute’s 18th Annual Monetary Conference cosponsored with The Economist on October 19, 2000.

Despite the overwhelming evidence that markets perform best when left alone by the government, it is still virtually taken for granted that one consumer product should be completely controlled by every government in the world. One product, so ubiquitous, that it’s used by almost everyone in the world on a daily basis: money.

Money is vitally important; the lifeblood of our financial system, but it is a product nonetheless. Consumers use this product not just as a medium of exchange but also as a liquid store of value and as a basis for accounting. Money producers, i.e., central banks, profit through seignorage, the ability to earn interest on their assets while issuing notes, e.g., dollar bills that pay no interest. The Fed creates money from thin air when it buys an interest bearing Treasury security and credits the seller with dollars. These dollars are not backed by anything in the sense that the Fed is not obligated to buy or convert dollars into anything.

Now there would be nothing wrong with this if the Fed were just some private institution trying to earn a living in an unregulated market. But of course, the Fed is a government-protected monopoly. Even prior to the Fed’s creation in 1913 there was always some level of government regulation of money. Prior to the Civil War, private banks used to issue dollar notes that were convertible into gold. Scholars debate whether the banking crises and panics during this period were a product of government regulation of these currencies and banks. What is clear is that there was never a completely free market in currency issuance. Well, the time has come.

To believe that the Fed is necessary is to believe that money is such a special product that it is optimal to give power to a group of expert economists either to use their best collective judgment in setting policy or to remove their discretion and create certain rules for them to administer. The only alternative is to eliminate government control of money.

Let’s examine these beliefs. Currently, Fed policy is essentially targeting low inflation. There are discussions about whether there is sufficient productivity growth to allow a high rate of GDP growth but the bottom line is protecting against an increase in inflation. While the target is not explicit, I would argue that markets are free enough and evolved enough to force the Fed towards this implicit target.

The Fed controls the spot interest rate market (Fed funds) by creating or destroying spot dollars. All the other points along the money market yield curve are more or less determined without government interference. If the market feels the Fed is being lax on inflation, all the non-spot money market rates will rise and spot dollars will be sold for inflation hedges like commodities and foreign currencies. These trends will continue until the market catches the Fed’s attention and the Fed funds rate target is increased.

A more efficient policy would be to announce an explicit inflation target. With the continued development of the inflation protected bond market, which trades on real rates, this policy could be accomplished by explicitly targeting the market’s expected inflation rate (a first order approximation is calculated by taking the geometric difference of real rates from the nominal rates of comparable bonds). The Fed could choose the means by which to affect the markets. But this still leaves the problem of what inflation rate to target and how to define an inflation index.

It is arguable whether stock and real estate prices should be included in an inflation index because price bubbles are destabilizing and an indication that monetary policy is too loose. There certainly are non-linear effects in markets that can cause self-reinforcing trends that push markets away from their fundamentals. But bubbles, at best, can only be defined after the fact. Like an Oliver Stone movie where lack of evidence is an indication of the strength of the conspiracy, prices continuing to go up are evidence of the increasing size of the bubble! The term is meaningless to describe current conditions because it can’t be disproved; prices go up because there is a bubble; prices go down because there was a bubble, either outcome is proof of the existence of a bubble. So let’s stick with CPI.

Changes in the CPI can occur for non-monetary reasons (i.e., events independent of too much or little money creation). If the Fed is going to target CPI and minimize inflation due to monetary policy it has to be able to adjust for these non-monetary events. Central bankers are happy (and correct) to overlook price rises from supply shocks, such as droughts. The flipside is that price drops due to supply shocks coming from productivity growth also need to be adjusted for. If some innovation allows the cost of producing widgets (of equal quality) to drop the credit should not go to monetary policy.

So the ability to measure productivity is critical regardless of whether an inflation target is explicit or implicit. Even with the way current Fed policy is formulated, productivity is at the core of the debate over the appropriateness of policy. Higher productivity growth means the economy is running more efficiently and can therefore run faster without increased inflation. But the technological and financial advances that we have seen and will continue to see (if the government can stay somewhat out of the way) are quickly transforming our ability to define let alone measure productivity.

In the economy of the 1950s productivity was easy to measure as it was largely a function of widgets produced per man-hour. Today more and more firms are valued not for how many widgets they produce but for the quality of the ideas they produce - their intellectual property. These ideas may not generate any revenue today but in the future, and can therefore only be calculated today by looking at a firm’s market capitalization, a highly volatile measure. (This is not to argue that a firm’s market value is a perfect measure but that a priori it is the best unbiased measure available.)

Consider a firm whose sole function is to develop patents for licensing. The only quantitative measure of their output will be the number of patent applications generated per man-hour. However, what is important is the economic value of these patents, which may not correlate with the number of patents generated. This economic value can only be ascertained by seeing what value the market gives to the company.

While there may not be a lot of pure research companies like this, these types of companies effectively reside within virtually every tech firm. The research aspect of a tech firm is perhaps its most vital part because the trade secrets and patents produced are the engine of its future growth.

Many would be happy to sweep these issues under the table because, in their view, the economy is doing great and the Fed deserves much of the credit. In other words, if it ain’t broke don’t fix it. After all, the stock market has had an incredible run (despite being down some this year), the dollar is strong, inflation is seen to be under control, and the last time unemployment was this low Bill Clinton was busy not inhaling. Going back to the start of 1995, the S&P 500 has more than tripled and GDP growth has been 4.1 percent (go back further and this rate drops).

But as good as things have been, they fall short in comparison with the average GDP growth of over 5 percent in the hundred years prior to 1971, when Nixon closed the gold window and floated the dollar. This may not sound like much of a difference but it means that national income would have been more than 25 percent higher during these past five years had the economy been growing at a historically average rate.

Additionally, while inflation is low, it isn’t zero (even considering any and all measurement biases). So our money is still being debased, albeit at a slow pace. Given the tremendous technological advances and the subsequent increases in productivity, we should really see falling prices if the dollar were truly maintaining its usefulness as a store of value. Not a 1930s-style severe deflation that was caused by a variety of bone-headed fiscal and monetary policies that squashed both demand and supply. Just a steady, modest price decline reflecting productivity growth adjusted for quality improvement in products and services, where even products whose quality stays the same become cheaper to produce, raising the standard of living for all.

This recent period has also had many advantages over the prior period. With the Cold War over and defense spending being cut, valuable capital and human resources have been deployed in far more productive activities. Financial markets have evolved to be far more efficient, sophisticated, and global than a generation ago, allowing capital to be allocated more efficiently.

U.S. corporate performance has reaped the benefits of the restructurings during the 1980s. Granting employee options is now prevalent. This gives both better incentives to workers and provides a tax-advantaged way of compensation resulting in a far more efficient way to pay staff.

Yet despite these advantages we can’t seem to match the growth rates of the past. Of course this foregone income cannot all be chalked up to the failings of monetary policy. Certainly the enormous increases in taxes, regulation and government spending bear much of the blame.

But a truly market-based monetary policy might not have been such a willing accomplice to the increasing encroachment of government into the private sector. Without the implicit inflation tax and monopoly profits from the Fed, government would be forced explicitly to raises taxes (politically difficult) or to lower spending and deregulate. If the latter path is accompanied by tax cuts, experience shows that the economy benefits. In other words, the dynamic of tighter monetary policy (to eliminate current inflation) hurting short-term growth can be completely offset by fiscal policy. (Shockingly, many still hold to the Keynesian notion that increasing spending is the fiscal antidote to slowing growth when, in fact, it will exacerbate poor economic conditions by displacing the market’s more efficient allocation of resources. Japan in the 1990s is the latest example of this although they suffer under many other bad policies too.)

There’s certainly reason to have confidence that competent Fed officials will be able to avoid a sustained resurgence of inflation. It is also clear, as shown above, that there are many reasons to believe that we should be doing better and that therefore monetary policy is not optimal. The simple fact is that it is not just hard but impossible to tell what exactly the right policy should be at any given point in time. Economies are just too dynamic and are composed of too many players. No matter how smart members of the Fed may be, it makes no more sense to have them set monetary policy that it did for elite economists in centrally planned economies to set the price of food, cars or any other product.

Even if it were possible for Fed members to have completely mastered the art of central banking, is it healthy to vest so much power in one person or group of people? Imagine how the markets would react if Greenspan should have a very sudden demise.

So, how do we solve this problem of imperfect people using imperfect data creating a one size fits all policy? The first step is to remove discretion over policy from the government. The explicit targeting of inflation, previously mentioned, accomplishes this except that setting the goal would still be discretionary and likely to be sub-optimal given productivity measurement problems. An alternative is a return to a gold standard as way to effectively automate monetary policy. Despite its popularity in some quarters there are serious problems with any type of gold standard.

Go ahead and ignore the fact that no country has ever been able to maintain a gold standard (perhaps with Hong Kong and Argentina both having kept their currencies fixed to the dollar even during periods of severe economic distress, it is an indication that times have changed). The main flaw in fixing the dollar price of gold is that it assumes that the value of gold doesn’t change, like some physical constant, as immutable as the maximum speed of light. Gold’s value comes in two parts. First as a commercial product with limited use as an industrial metal and as jewelry or ornamentation. Clearly, the price of any commercial product will vary for a variety of reasons having nothing to do with an economy’s general price level.

The other component of gold’s value derives from its long history as a store of value and is wrongly assumed to be intrinsic. The supply of gold varies as new mines are found and mining techniques improved. Demand for gold as a store of value is based on the ability of gold to compete against other liquid stores of value. So, the only advantage of gold over a fiat currency is that the production of gold is not controlled by the government (ignoring taxes and regulation).

But if we fix the dollar to something that floats are we accomplishing anything? If the supply of gold increases due to a new way to mine, the dollar price of gold needs to fall. (As an example if the supply of gold doubled as a result of productivity boosts and there was no change in the productivity of producing other goods, the dollar price of gold should be halved as to protect the purchasing power of the dollar.) If the dollar price of gold is held constant, as in the gold standard, inflation will ensue (which is historically what happened under these conditions).

Irving Fisher attempted to solve this problem with his compensated dollar plan. The plan allows for the dollar price of gold to be adjusted by a CPI priced in gold. The problem is that this plan only works if productivity in the mining sector is the same as in all other sectors of the economy. Not a bad assumption, perhaps, in Fisher’s day but unlikely to be true now and even less likely in the future. Two other possible solutions are to let the price of gold vary based on the relative productivity of gold production to productivity in the rest of the economy or instead of fixing the price of gold, fix the dollar price of gold plus other commodities (i.e., define the dollar as some fixed basket of commodities).

Well, the first solution is pretty messy, as it would require the government to calculate productivity rates that as argued above are inherently impossible to measure precisely. Furthermore, even if productivity could be measured precisely, it could in practice only be measured with a time lag. Not very helpful for knowing today’s price of something.

The second solution sounds more workable but determining the composition of the basket that would define the dollar is tricky. Leave something out of the basket and you have the same problem of the relative production productivities of what’s in and out of the basket. So just about everything the economy produces must go in the basket (obviously there are diminishing returns to accuracy for each additional product added, so you start with the most important products and work towards things that are used less). Now with many products in your basket you need to determine relative weights to reflect relative usage and importance of these products. But the resultant basket is just the same as that used to calculate inflation (with the identical problems as mentioned earlier)! Linking the dollar to this basket is exactly the same as having the Fed explicitly target zero inflation.

The key to getting better monetary policy is not to merely limit the discretion of the government but to get the government out of the money business altogether by privatizing the Fed. Sell the whole thing to the highest bidder. The government would also have to deregulate enough to allow competitors to arise in the currency issuance game.

By eliminating this government protected monopoly, more of the informational value of consumers ever changing preferences and behaviors could be used by producers to create a product that can best balance everyone’s needs. The free market can be looked at as a computer, calculating within all the constraints and using the near infinite amount of interrelationships and feedback between economic actors, on a global scale, to solve the problem of what are the best forms of money. This is an impossible calculation to do any other way.

Consumers would choose between U.S. dollars, euros, Citi dollars, GE dollars, etc. This choice would be based on confidence in the issuer and how well the product serves the consumer’s needs. Companies would issue money solely as a means to profit. Produce too much money and it becomes worthless, too little and not enough people will be able to use your money for you to profit. The notion of a government having a monetary policy goes the way of governmental industrial policy (or choosing which firms receive government help - see Asia).

If private firms were allowed to compete on equal legal footing as a private Fed, currency competition would lead to better money just as market forces improve the quality of any product. In an unfettered environment, using precious metals as a backing for a currency is just a starting point. Nobody can predict the improvements and ingenious ideas that would emerge. Already consumers have reaped benefits from quasi-currencies like airline miles and credit card rebates in the form of products (clearly, these things don’t currently have the liquidity to make for a good form of money).

There are usually several objections to private money. It is argued that it is necessary to have a lender of last resort in times of crisis and that only the Fed can fill this role. Of course, there is no reason why a private central bank couldn’t provide sufficient liquidity during a crisis and no reason why private regulators couldn’t get big financial institutions together to provide liquidity (J.P. Morgan did this in the panic of 1907, prior to the Fed’s creation). Also prior to the Fed’s creation, private clearinghouses would lend to members who were solvent but needed liquidity. Furthermore, using the Fed as a lender of last resort creates moral hazard. If the perception is that the Fed will be compelled to act in a crisis, the effect will be to allow participants to take more risk, as they believe they are receiving some downside protection from the Fed.

In 1998, a financial crisis that started in Asia threatened world markets. “Contagion” was used to describe the effect of countries with healthy economies seeing their markets roiled. The contagion spread through trade and capital flows between countries and from the market realizing countries with a certain economic profile (high current account deficits, low foreign exchange reserves, and a pegged currency) were vulnerable. Markets in developed countries became infected as credit spreads widened and volatility increased (a mathematical implication being that correlations between markets increase). To extend the biology analogy, the best defense for a population against mutant viruses is genetic diversity. The economic equivalent is the diversity of products and regulation that spring forth from a free market. This diversity would reduce the frequency of crises that might need a lender of last resort.

Another concern about a world with only private money is that things would be too complicated with all the exchange rates that would exist for these new currencies. Technology can easily solve this problem. Only want to see prices in one particular currency? Your handheld device or browser could automatically convert all prices and even facilitate conversion of your currency to one that would be acceptable to a merchant. Currencies that were too volatile would quickly fall into disuse, as part of their utility would always derive from them being a stable store of value.

As with any free market reform, there is no expectation that private money would lead to a perfect world where there are no crises or problems, just a better world. Better not just in a strictly utilitarian sense but also in a moral sense as people could store the fruits of their labor however they see fit and not be forced to submit to a tax (inflation) that is not explicitly levied and voted on.

Many involved in the information revolution are confident that with the help of new technologies, markets will continue to evolve and reach a point that government’s ability to tax and regulate will be stifled. What these optimists miss is that government too has the ability to evolve and counter these trends. Government’s monopoly on the use of force can trump any market innovation. Ultimately, it is necessary for the political climate to change before government will acquiesce and not try to fight the liberalizing effects of technology.

These changes are certainly not around the corner, but they should be our goals.

  • Repeal all laws giving the Fed monopoly power to issue currency.
  • Eliminate all government regulation of banks and financial institutions.
  • Change the tax laws so there was no preference given to transactions done in dollars (payment could still be required in dollars but there should be no capital gains if you choose to hold your money in yen, for example).
  • Completely privatize the Fed either through an auction or fair distribution of shares to taxpayers.
  • Exempt banks and financial institutions from antitrust laws.
Deregulate. Privatize. To improve our standard of living, to increase the level of freedom, and to set a powerful example for the world to follow.
Bob Gelfond is CEO and Founder of MagiQ Technologies. This article was published in the Cato Journal, Fall 2001.