Hidden Biases Determine How the Fed Sets Interest Rates
The basic mandate of the Federal Reserve is to promote broad economic conditions to achieve maximum employment and price stability. Since its mission focuses solely on macroeconomic conditions, monetary policy is considered to be unbiased.
Yet, can monetary policy be unbiased from its mandated objectives, but biased in the way it attempts to achieve them? While the mandate of the Fed has been straightforward and constant, the operating framework in which officials try to meet their objectives has indeed changed.
For example, for much of the postwar period the implementation of policy focused on setting parameters for expansion in money supply and private credit. Depending on the economic environment, growth rates for money and credit were adjusted up or down and policy makers would announce targets annually. Deviations would sometimes trigger a change in official interest rates, especially if the economy and inflation weren't performing in line with expectations.
But financial innovations created new instruments like mutual funds, which enabled people to move money back and forth from checking to saving. These new instruments lived outside the banking system and therefore weren't captured in traditional measures of money. In time, this altered the relationship between money and economic activity in such a way that the Fed abandoned money supply as a policy tool in the early 1990s.
The Fed then shifted to focus on real interest rates, which were determined by subtracting the inflation rate from nominal interest rates. That proved to be workable for a while, but policy makers soon found problems here as well since the method was backward looking and the hallmark of a successful monetary policy focused on stabilizing future inflation.
In the 2000s, the Fed started to follow an inflation-targeting regime, but it was based on discretion rather than a specific price rule. That approach lasted until 2012, when the Fed explicitly adopted a price rule of 2 percent a year in consumer prices.
Over the past several decades there have been four basic operating frameworks of the Fed; growth in money and credit, real interest rates, and both a discretionary rule on inflation and a formal one. Does any one of these have greater potential to create a bias or uneven distributional effects in the economy?
Policy frameworks that focused on broad money and credit growth as well as real interest rates would be free of any economic bias since policy makers are setting parameters for the entire economy and not any specific sector or industry. Also, money and interest rates are something the Fed controls directly or indirectly.
However, price-setting targets cross the line in a way as the Fed is getting directly involved in decision making of the real economy and, as a result, price-targeting regimes have the potential for creating biases or unintended micro consequences.
To be fair, merely by targeting a measure like the consumer price index to grow at a specific rate by itself is not creating a bias in the economy. Indeed, policy makers are setting a macro rate of inflation and it's up to the marketplace of consumers to determine how that inflation is divided up.
Yet, the potential bias could come from what’s not included in the price-targeting regime. For example, consumer price indexes do not include housing prices, though for decades house prices were directly part of the CPI and, until the late 1990s, owner-occupied housing was part of the rent sample. Financial asset prices are also not included, but many think these prices should be part of a consumer price index that attempts to measure the cost of living . Has their exclusion led to any uneven distributional effects in the economy? I can think of at least three important ones.
First, house price inflation used to run more or less in line with consumer price inflation. Yet, now that central banks are targeting general inflation and house prices do not directly or indirectly impact the targeted inflation index, the relationship between house price and general inflation has become unhinged. In the 2000s economic cycle house prices ran three to four times faster than general inflation and in the current cycle house prices are running two to three times faster.
Second, overall asset inflation, and, in turn, wealth creation, is markedly stronger since the shift to price-targeting regimes. That is most evident in the household net-worth-to-income ratio. Until the late 1990s this most important ratio moved in a narrow band, with the level of household wealth averaging roughly five times more than income. Yet, in the last 20 years, due to relatively fast and persistent increases in real and financial asset prices, this ratio of wealth to income has reached higher and higher highs, only to fall back sharply when asset price bubbles burst. Based on actual asset prices, household wealth will stand at nearly seven times income at the end of 2017
Third, the share of corporate profits to GDP has increased by roughly 3 percentage points since the informal and formal price-targeting regime was implemented. Two of the 3 percentage points are accounted for by the finance industry, a sector that clearly benefits from financial asset price inflation.
None of these examples prove causation. But it is hard to arrive at that conclusion as the shift to a price-targeting regime means the relative out-performance of asset prices is a mere coincidence. Indeed, recent asset price patterns raise serious questions about whether the current price-targeting framework is in fact biased, lifting real and financial asset prices far above what otherwise would occur if a different monetary policy framework was employed. And what elevates the potential bias even more so is that policy makers have in the past reacted quickly and dramatically to major reversals in asset prices yet avoided making any policy changes when asset inflation has been rampant.
Discussions are now underway for policy makers to consider a new operating framework. Those who continue to defend the current framework fail to acknowledge the potential biases of the inflation-targeting regime and, most importantly, this framework hasn't generated better macroeconomic performance and stability. Indeed, two recessions have occurred since formal and informal price targeting became been a key policy tool (the second was the most severe since 1945).
One proposed fix is to increase the targeted rate of inflation from 2 percent to, say, 3 percent or 4 percent. The thinking is to let the economy run hot so there is a bigger cushion on inflation and interest rates. This proposal is highly contentious, but it should be noted that there is no empirical research that ever proved 2 percent is the right amount of inflation.
The main problem with the price-targeting regime is not the framework, but the measure of prices that are targeted. Consumer price series are not a pure measure of inflation as they include a number of imputed prices (e.g. owner-occupied housing) that are not relevant or timely for the conduct of monetary policy. The only way a price-targeting regime can work is if the targeted series is capturing a broad and accurate measure of actual inflation in the economy.
Creating a broader transaction-based price index, which at least included house prices, wouldn't be such a novel idea since house prices were part of the original consumer price index. Restoring them would give policy makers a more direct link to the assets that are highly sensitive to interest rates and changes in monetary policy. No single fix would eliminate all of the potential bias of the current framework, but it would better enable policy makers to track the changing ways monetary policy is affecting the economy. While they are at it, it might also be prudent to raise the price target.
Professionals offering actionable insights on markets, the economy and monetary policy. Contributors may have a stake in the areas they write about.
To contact the editor responsible for this story:
Daniel Moss at email@example.com