For by grace you have been saved through faith. And this is not your own doing; it is the gift of God, not a result of works, so that no one may boast.
—Ephesians 2:8–9
The natural interest rate was the most important variable of monetary economics in the twentieth century. It has lingered in the background of economic policymaking, dictating the terms of debate for how economic policy in general, and central bank policy in particular, is conducted. For the past thirty or so years, it has been behind the decisions of central banks to raise and lower interest rates—policies that can generate recessions and create mass unemployment, and thus have an enormous impact on everyday people.
In 1517, Nicolaus Copernicus wrote a memorandum on the monetary policy of the Kingdom of Prussia. Price inflation, he argued, was caused by an excess of money in the economic system:
Money can lose its value also through excessive abundance, if so much silver is coined as to heighten people’s desire for silver bullion. For in this way the coinage’s estimation vanishes when it cannot buy as much silver as the money itself contains, and then I find greater advantage in destroying the coin by melting the silver. The solution is to mint no more coinage until it recovers its par value.1
At roughly the same time, the School of Salamanca, led by Martín de Azpilcueta, formulated a consistent theory of the relationship between prices and money, arguing that it was the influx of gold and silver into Spain that caused the rise in prices—the so-called price revolution of the sixteenth century.2
The theory is intuitively appealing. Imagine there are ten identical goods in the economy. Now imagine that there are ten people with identical preferences who each have one dollar—the money supply is thus ten dollars. We assume that all ten people spend their whole income. Each good will then be worth one dollar. If the money supply is increased to, say, fifteen dollars, the dollars are distributed evenly, preferences stay fixed, and all income is spent, then the price of each good rises to one dollar and fifty cents. Now assume that each person saves ten cents; this reduces the price of each good to one dollar and forty cents.
The most popular formalization of this theory was undertaken by Irving Fisher in the early twentieth century. Fisher argued that the key intuition behind the theory could be captured by
$M.V = P.Q,$
where M is the supply of money, and V is the velocity—meaning how quickly the money is spent or, alternatively, the turnover of the money supply in a given period. P is the level of prices, and Q is the quantity of goods sold.3 A mathematical identity, it is true by definition. But quantity theorists made a causal claim when they rewrote the equation as
$M.\bar V = P.\bar Q.$
The symbols $\bar V$ and $\bar Q$ now designate fixed quantities, and the equation is meant to be read from left to right: if the money supply is increased, the price level increases with it, just as Copernicus and the School of Salamanca suggested.
Fisher having made his case, problems with his theory soon became apparent. For one thing, the quantity of goods sold did not appear to be constant. The business cycle revolves once in a decade or so, resulting in large fluctuations in the quantity of goods sold. Nor was there reason to assume that the velocity of money must be constant. Cash may be spent or set aside, and the ebbs and flows in consumer spending lead to an ever-changing and often discontinuous velocity.
Most damning were the critiques of the Banking School in Britain. Led by Thomas Tooke, a London banker who pioneered the study of monetary statistics, the Banking School argued that the causal arrow in the second equation should run from prices to money and not money to prices.4 Their argument was based on observation: prices rose before the supply of money changed.
If so, why?
The Banking School argued that money creation could not be externally controlled by any authority and was, in fact, determined by the amount of lending that banks undertook. In a world in which money was first and foremost an entry on a banking balance sheet, the creation of money was determined by (a) the amount of money the banks were willing to lend at a given price, and (b) the amount of money borrowers were willing to borrow. The Banking School argued that when the economy started to run hot and prices started to rise, the banking system would accommodate this increase by increasing the amount of money in circulation.
The quantity of money is an endogenous variable, controlled from within the system itself, and for this reason analytically useless. Many late nineteenth-century economists thus turned their attention away from the quantity of money and toward its price.
It is the interest rate that reflects money’s price and so the value placed on its use.
The first economist to investigate interest seriously was Knut Wicksell. In Interest and Prices, Wicksell is quite clear that his views represent a logical progression from the quantity theory of money:
The Quantity Theory … is open to too many objections, as pointed out by later writers, to be accepted without modification. The only possible course seemed to me to attempt to push on … to follow up the logical consequences of the fundamental conception that had given rise to the Quantity Theory, so as to arrive at a theory which should be both self-consistent and in full agreement with the facts.5
Wicksell came to distinguish between two different rates of interest. One was the money rate of interest. This is the rate that bankers charge borrowers and, as Wicksell observed, it is subject to fluctuations based on their propensity to lend. The other rate of interest Wicksell called the natural rate of interest.
In the quantity theory, if banks increased the money supply an increase in prices would follow. But Wicksell thought that a price hike in such circumstances was not primarily due to an excess of money; rather it was because money was being offered too cheaply—at too low a rate of interest. This encouraged overborrowing, which then bid up prices. The question then became, “Too cheap relative to what?” Wicksell claimed the money was too cheap relative to the natural rate of interest. And the natural rate of interest, he went on to argue, was equal to the rate of profit on capital goods. If the rate of profit is on average six percent, the natural rate of interest is six percent. If banks are willing to lend money at four percent, what then? Anyone considering investing in capital goods would obviously take advantage of the spread, borrowing at four percent to invest for a return of six percent.
In finance, this is called an arbitrage; and, in life, a very good deal.
Good deals create problems all their own. Money that is borrowed with an eye toward interest rate arbitrages creates an investment boom. Money pours out of the banks, as investors buy machines and buildings to set up new factories and hire more people to operate those machines. If all the machines, buildings, and people are already being put to work, this will tend to bid up the prices of machines, buildings, and wages. As wages rise, so, too, consumption, and so, too, prices of consumer goods. This trajectory introduces the quantity theory through the backdoor. Too much money is now chasing too few goods because the banks were willing to set the money rate of interest below the natural rate of interest. In mathematical terms,
$f\left( M \right) = n - i,$
where M remains the money supply, n is the natural rate of interest, and i is the money rate of interest. If
$M.\bar V = P.\bar Q,$
it now becomes clear that
$n > i = \Delta {M^ + } = \Delta {P^ + }.$
This theory proved seductive in its simplicity. Other economists soon used it to explain the business cycle. Friedrich Hayek and Ludwig von Mises both noted that during the upswing of the business cycle prices tended to rise, while during the downswing, to fall. They reasoned that during the upswing the money rate of interest must fall below the natural rate, while rising above the natural rate during the downswing. The business cycle became a financial drama, with a boom in the first act as bankers sold loans cheaply, and a bust in the last act as prices spiraled out of control.6
Other economists, less interested in the drama and more interested in economic policy, were quick to see the Wicksellian conceptual framework as being a perfect means by which central banks could attempt to steer the capitalist economic cycle. In his early work on monetary economics, John Maynard Keynes showed how the central bank could control the rate of interest that banks charged the general public—the money rate of interest. An alignment of the money rate of interest and the natural rate of interest would give rise to stable economic growth, with little in the way of either inflation or deflation. Keynes was a man of experience, and he did not allow himself to get carried away. At the end of his Treatise on Money, he wrote:
[T]he mere occurrence of a Credit Cycle is in itself a demonstration of the fact that the Banking System has failed to change the market-rate so as to keep pace with changes in the natural-rate. It is certain, therefore, that hitherto the Banking System has not succeeded in controlling the Rate of Investment with sufficient success to avoid serious instability.
Thus we cannot do more at present than marshal the various means at the disposal of the Banking System. Only the future can show for certain whether the conscious and well-directed use of all these means, confidently employed in the right degree and at the right time, is capable of solving the problem.7
This would be the refrain of economists for the next century, even though Keynes himself soon changed his views. It is no surprise that he did so. Wicksell’s views very soon came under fire. The year after Keynes’s book was published, John Williams wrote:
It sounds like a solution of the difficulty, but amounts merely to another way of stating the difficulty. If the natural rate were visible, the case might be different, but only the market rate is known. The natural rate is an abstraction; like faith, it is seen by its works. One can only say that if the bank policy succeeds in stabilizing prices, the bank rate must have been brought into proper line with the natural rate, but if it does not it must not have been.8
Williams was onto something, but the difficulties were not just epistemic. There were other serious problems with Wicksell’s theory.
Economists throughout the twentieth century and into the twenty-first have attempted to reduce Wicksell’s theory to a set of actionable rules. None has been as successful as John Taylor, who has used the natural rate framework to create a monetary policy rule for central banks.
The standard version of the Taylor rule is
$r = \pi + g.y + h\left( {\pi + {\pi ^*}} \right) + {r^f},$
where $r$ is the central bank policy rate of interest, $\pi $ is the current inflation rate, ${\pi ^*}$ is the equilibrium or target rate of inflation, $y$ is the difference between real GDP growth and potential GDP growth, and rf is the natural rate of interest. The terms $g$ and $h$ are coefficients that depend upon the person using the rule.
Taylor rules are today calculated by central banks. They are not typically used to set policy; but they are studied by central bank economists, who, if skeptical of the estimates that they require, nevertheless believe in the framework that they impose.
What is odd in all this is the indifference of so many practical people to the theory’s empirical content. Interest rates are said to affect the economy by increasing or decreasing the rate of business investment. Yet when actual data on interest rates and private sector investment are collected, no relationship is forthcoming. Why is it that? Surely the economic theorists have a point, and the rate of borrowing should affect the rate of investment.
When the money rate of interest is equal to the natural rate of interest—which is the return on physical capital, plant, and machines—there is supposed to be a so-called correct amount of investment. This correct rate of investment ensures that the economy is not subject to inflation or deflation, that there is not too much spending or too little, that it runs neither hot nor cold—but just right.
One immediate problem: not all investment is created equal. Too much bad investment that does not lead to the future production of goods and instead leads to mass bankruptcies might lead to inflation.
In the 1930s and 1940s, the economists Roy Harrod and Evsey Domar noticed that investment is double-edged. When investment is undertaken, the immediate result is an increase in spending and so an increase in aggregate demand. An increase in spending can thus lead to an increase in prices. But investment increases the potential output of the economy and thus increases aggregate supply. Such is the Harrod–Domar knife-edge.
The knife-edge has generated many debates.9 Proponents of the natural rate of interest assume that the economy will automatically factor the correct increase in output into investment; this, in turn, will be priced into the interest rate.10 The economy is the sum of individual decision makers; and the decision makers that determine the amount of investment are those that run the firms. On average, it is implicitly assumed that firm management will favor a rate of investment that generates the level of earnings they expect—the rate of profit, the natural rate of interest. All should be well.
But the empirical literature shows that management tends toward overconfidence in future earnings and that this affects their investment decisions. One recent summary:
A large and growing body of evidence suggests that a substantial share of top corporate executives exhibit symptoms of overconfidence in their decisions … The presence of CEO overconfidence … seems to matter for a variety of firm decisions and the choice of financing for those decisions. Notably, it matters for the extent to which investment choices, both those involving internal investment and external mergers, track the available cash and easy-to-obtain debt available to firms.11
What is more, this overconfidence is not static. It appears to go through boom-and-bust cycles. This can be seen in the dot-com investment and stock bubble of the late 1990s, when CEOs became four and a half times more overconfident than they had been before or have been since!12 Investment poured into hundreds of internet firms that would prove unviable. In 1999 alone, 457 companies were launched onto publicly traded indices, and, of those, 117 doubled in price on the first day of trading.13 In the United States in 2000, around one hundred billion dollars of venture capital investments were undertaken. This figure was up from around eight billion dollars in 1995 and would fall to twenty-one billion dollars in 2002.14
There seems to be no reason to assume that, if the banking system manages to align the money rate of interest with the natural rate, entrepreneurs and firm management will respond rationally and invest as they should. It seems just as likely that they will be subject to swings of overconfidence.
The assumption that there is a single interest rate at which firms can borrow is common in the literature devoted to the natural rate of interest. That rate is set by the central bank. But the central bank is not the only player. When financial markets lend to firms, they do so in the corporate lending market. There is a very long-term relationship between the central bank rate and the corporate lending rate, but nothing in the short or medium term, and the spread between rates varies quite a lot.
Market overconfidence is easy to study because stock analysts regularly publish their expectations. Actual earnings growth can vary widely from expectations.15 Financial market analysts, on average, forecast earnings growth at approximately thirteen percent a year. But the actual average earnings growth is around seven percent a year. In short, financial markets seem to think that firms will have roughly double the earnings growth that they actually experience.
It looks as if markets are not very good at pricing future outcomes. Since CEOs seem prone to overconfident mood swings too, it is hard to see any evidence that there is a channel from the money rate of interest, through the speculative markets, via the overconfident CEOs, all the way to the optimal rate of investment that keeps the economy running just right. Given the number of moving parts, the chance that an optimal rate of investment would arise in response to even a correctly estimated natural rate seems extraordinarily slim.
If economists were to jettison the idea of the natural rate of interest, where would they be? Central banks still must set the interest rate, after all. Without the natural rate of interest, what alternative benchmark could be used to control prices?
Perhaps economists should go right back to basics and think of the rate of interest as the classical economists did until the nineteenth century: as a distributional variable. From this perspective, the interest rate is simply the rate of increase of purchasing power held by those who hold savings in risk-free assets. This contrasts with both the rate of increase of purchasing power of those who own physical capital—the rate of profit—and the rate of increase of purchasing power of those who sell their labor—the rate of change of wages.
These three variables overlap with three social classes: rentiers, capitalists, and workers. These should not be thought of as distinct people or agents. Rather they are functions of their income sources. A person may be a worker when receiving wages and a rentier when accumulating a pension via the stock and bond markets. The question of where to set the interest rate then becomes a distributional matter: how much should rentiers’ income increase every year relative to the income of capitalists and workers?
The economist Luigi Pasinetti responded by defining a fair rate of interest, one stemming “from the principle that all individuals, when they engage in credit/debit relations, should obtain, at any time, an amount of purchasing power that is constant in terms of labor.”16 The idea is that the rate of increase of income for every class—rentier, capitalist, worker—should be tied to the productivity of the worker, that is, to the real wealth that the economy is able to produce given today’s level of technological development. Holding these rates constant to one another ensures that the income generated by the economy is distributed evenly. Setting the interest-rate policy rule then becomes simple:
${i_{fair}} = \hat y + \hat p,$
where ifair is the fair rate of interest, $\hat y$ is the growth rate of labor productivity, and $\hat p$ is the rate of price inflation.17 This rule is quite like the old Scholastic notion of a fair price. Marc Lavoie points this out when he writes:
Thus, in a world with no technical progress and no inflation, the nominal interest rate ought to be zero, as was argued by the Church at the time [i.e., the Middle Ages] when these conditions were roughly fulfilled.18
In other words, usury is only to be tolerated as a means to distribute some constant share of the growing economic pie to savers. This share is kept constant by assuming that it—together with wages and profits—only grows in line with the technical innovations that allow for more productive industry. The Scholastics may have had a cruder understanding of the macroeconomy than possessed today, but their instincts were correct. If it looks like a duck and quacks like a duck, then it is probably a duck. And if it looks like a yield on rentier assets and is spent like a yield on rentier assets, then it is probably a yield on rentier assets.
Unlike the natural rate of interest, the fair rate is measurable by means of statistics that are in widespread use today. It is visible. And it is not aimed at trying to manipulate economic activity—a task that should be done by the government through Keynesian fiscal expenditure and taxation. The interest rate should be recognized for what it is: the yield of rentier assets.
The natural rate theory demands faith in an unseen entity: the natural rate of interest. Its proponents in and around the central banks assure that if economists have enough faith in this entity and try our best to accommodate our banking system to it, we will be the recipients of economic grace. It is an eminently Protestant theory.
The alternative is that we see the interest rate for what it is and judge it by its works: the amount of money distributed to rentiers. We then—like the pre-Reformation Church—judge the interest rate according to its overall distributional fairness.19
What good is it, my brothers, if someone says he has faith but does not have works? Can that faith save him? If a brother or sister is poorly clothed and lacking in daily food, and one of you says to them, “Go in peace, be warmed and filled,” without giving them the things needed for the body, what good is that? So also faith by itself, if it does not have works, is dead.
—James 2:14–17