Monthly Archives: December 2010

Foolish Fed Policy Recommendation

Scott Sumner, writing here in National Review Online, proposes that the Fed focus on targeting Nominal GDP (NGDP) because:

  • Deflation is very, very bad (viz. Japan since 1990)
  • Government inflation measures (i.e., CPI) are subject to manipulation by government statisticians and difficult to measure accurately
  • The fiat money system is here to stay
  • Gold-based monetary systems stabilize the price of gold and allow all other prices to fluctuate

Scott is off base for a number of reasons.

He is merely changing the output measure to target. Instead of measuring the difficult-to-measure Consumer Price Index, he proposes we measure the slightly-less-difficult-to-measure Nominal Gross Domestic Product. Both data series are designed by government, modified by government, and measured by government. Scott admits that one flaw in the current targeting approach is that the target measure (inflation rates) is subject to gamesmanship by the government bureaucrats. That is not just one flaw, that is the canary in the coal mine regarding centralized government planning or targeting of anything. The men and women that constitute the government planners and measurers are no better equipped than you and me when it comes to planning and measuring an economy as large and complex as ours. Any error, made because of malfeasance or incompetence, is grossly magnified because everything monetary goes through a single clearing desk: the Federal Reserve Board of Governors. Using outcome data, the Fed claims the ability to target (and the implication is to deliver) specific levels of economic growth in the future. The truth is the Fed does not know the future any better than you know it.

He is not acknowledging the fundamental reality of economics: output comes from increases in input and productivity. We only know the scale of productivity gains when we measure the output against the number of inputs (e.g., laborers). No better than the fellow who skips dinner three nights a week to save money for his invention in the basement that will change the manufacture of widgets forever does the government know which particular activities are heralds of a future increase in productivity. Even in a primitive economy, sources of incredible productivity gains had their origins in “foolish” endeavors. Consider the person who got tired of digging with a stone and started looking for something stronger that could be more easily worked into usable shapes. How he and his family must have been mocked by the community over the generations it took to search for bronze, figure out how to fashion bronze, and then demonstrate the utility of bronze. Central planners are in fact misnomers, they arrive to plan after the fact of individual creativity and productivity has been made evident. They are a drag on growth rather than an enabler of growth.

Scott does not seem to understand the order of economic decision-making. He says about the recent decline in nominal GDP and the debt crisis:

In 2009, the U.S. saw the biggest fall in nominal GDP (NGDP) since 1938. It is thus no surprise that we had a debt crisis: Borrowers almost always have trouble repaying debts when nominal income comes in much lower than was expected when the debts were contracted.

The debt crisis preceded the fall in nominal GDP. The increase in nominal GDP in the years before the debt crisis were in part attributable to the central bank’s efforts to impede the market’s natural function of pricing debt. In a fiat world, the monetary/political decisions made by the central bank and the fiscal/political decisions made by the legislature send signals to market participants about the price of risk, as reflected in interest rates, which help individuals decide whether or not to defer spending of their accumulated savings. When the central bank suppressed interest rates and the government agencies established to promote home ownership accepted low-quality loans into their guaranteed mortgage portfolios, individual investors received two signals for immediate consumption (a large part of nominal or real GDP) and residential construction investment (another large part of GDP). Bit by bit, marginal investors saw through the charade, until a tipping point was reached and the whole house of cards came crashing down. It was only after a year of malinvestment bills coming due that the government central bankers were spurred to act to resolve the “crisis.” Thus we had in late 2008 the spectacle of the Wall Street Secretary of the Treasury down on his knees begging an overmatched Speaker of the House to save the economy. (When the Speaker of the House explains repeatedly that unemployment is the most effective demand multiplier available, it is an act of charity to call her overmatched. To the extent she was selected because of her sex, let us all be reminded that affirmative action has a cost no matter how noble the intent may be.)

Scott moves to the question of the appropriate target for his preferred measure, nominal GDP. He notes Freidrich Hayek, who taught at Scott’s alma mater, U. of Chicago, proposed a target NGDP rate of zero. Hayek, brought up in the Austrian school of economics, knew that printing money was the starting point for boom-and-bust business cycles. The Great Depression was caused by this money printing in the United States after the creation of the Federal Reserve system and the vast expansion of consumer credit that spurred growth beyond a sustainable rate. That the responses to the Great Depression were wrongheaded does not detract from the reality that the source of the Great Depression were malinvestment booms lengthened by the fact that the government could print money longer without repercussions than could individual banks and other private lenders. The Japanese problems of the 1990s were the result of another long period of government-led malinvestment. The wonder of MITI, which was nothing more than mercantilism, paired with the Japanese investor community’s parochial acceptance of the low savings rates offered by the Postal System, created an artificially low interest rate that signalled savers to make over-investments in real estate and through leverage. We have not witnessed a decade and a half of deflation in Japan as much as we have witnessed a decade and a half of unwillingness to write off bad debts. The Japanese government has been propping up weak banks for far too long, all to the detriment of the Japanese economy. To the extent that the US government policy props up weak banks, we are repeating the Japanese “mistake.”

We can never eliminate error from human planning, so we will never eliminate booms and busts. What we can do is reduce the impact by moving the planning down to the lowest level possible and restoring at that lowest level the costs of error. When banks make loans that their loan officers know are unlikely to be repaid in full, they should suffer the consequences of those bad loans. That is not the case today. The old method of enforcing prudence on bankers was the bank run. We had one when Bear Stearns went under, and we should have let the other weak banks fail for their poor decisions. Instead, we propped them up, requiring strong banks to take TARP money to cover up the mistakes made by the weak ones.

Scott’s error is in trusting that central planners can improve economic performance. The gold standard, which he dismisses, is only the best standard because all the other possibilities are worse. Monetary systems have been abused since we had monetary systems: each time the metallic currency was recalled to stamp the new ruler’s face on the coins there was an opportunity to debase the currency, and the kings often took that opportunity. We have Gresham’s Law in economics (“the bad money chases out the good”) because of this tendency. The fiat money system makes debasement easier, which is why we have had such violent and global economic gyrations since the adoption of fiat money systems in the early 1900s. It doesn’t matter which output the fiat money governors use to measure because it is their system that is causing the trouble.