MISH'S
Global Economic
Trend Analysis

Recent Posts

Taxpayer Friendly Sites

Alphabetical Links

Thursday, March 03, 2011 2:40 PM


Goldman's Blood-Sucking Leeches Model, Money Multipliers, Macroeconomic Dark Ages, the Taylor Rule, and Nonsense from Trichet


Caroline Baum has an excellent column on Bloomberg today regarding money multipliers and a Goldman Sachs projection of what Republican budget cuts may do to the economy.

There were so many things in her post I wanted to reference that I asked Caroline if I could use her entire post. She graciously replied "Let 'er rip".

Goldman's Model Evokes Blood-Sucking Leeches: Caroline Baum

Macroeconomics really is stuck in the Dark Ages.

Take “fiscal stimulus,” for example, the idea that the government can step in to fill the void when the private sector isn’t spending and boost economic growth in the process.

Economists have been debating the pros and cons of fiscal stimulus since the 1930s, when John Maynard Keynes diagnosed the problem as one of inadequate private investment and prescribed public spending, financed by borrowing, as the cure.

The discussion hasn’t advanced very much in eight decades. Sure, economists have devised elegant mathematical models that purport to show that $1 of government purchases translates into -- take your pick -- no increase in gross domestic product (the multiplier is zero, according to Harvard’s Robert Barro) or $1.50 of GDP (a multiplier of 1.5, according to Berkeley’s Christina Romer, who was chairman of President Obama’s Council of Economic Advisers when the $814 billion stimulus was crafted in 2009). They haven’t really proven anything.

Keynesian economics went into hibernation in the latter part of the 20th century following an array of stimulus failures on the part of both Democratic and Republican administrations in the 1970s. The only thing the spending stimulated was stagflation.

In the 1980s, inflation came down, the Berlin Wall came down, economists thought the volatility of the business cycle had come down, and the notion of government as the solution went out of vogue.
Keynesians All

All it took was a good financial crisis for the Keynesians to come out of the woodwork.

The debate over fiscal stimulus went viral last week (at least in the geek world) with an economic forecast from Goldman Sachs Group Inc. (GS), a counter from Stanford University economist John Taylor (he of the Taylor rule), and an addenda from Goldman yesterday.

The Goldman gang projected an economic drag (that would be the opposite of stimulus) on GDP growth of 1.5 to 2 percentage points in the second and third quarters if House-passed budget cuts of $61 billion for the remainder of fiscal 2011 become the law of the land.

Asked about the Goldman forecast Tuesday following testimony to the Senate Banking Committee, Federal Reserve Chairman Ben Bernanke demurred.

“Our analysis doesn’t get a number quite like that,” he said. “Two percent is an enormous effect.”

He could have added: “especially when the rest of government is growing.”
Wrong on Everything

“Total government spending is up 6.7 percent in 2011 from 2010,” Taylor told me in a telephone interview.

Defense spending is rising, as are non-discretionary outlays for programs such as Medicare and Social Security that are on automatic pilot.

The proposed cuts would reduce non-defense non-security discretionary spending, a teensy share of the federal budget, back to 2008 levels.

In a Feb. 28 blog post, Taylor said Goldman’s analysis was “wrong.” He criticized it for failing to consider the beneficial effects that expectations of lower future deficits and smaller tax increases would have on the economy. He criticized the methodology for relying on the same “large multiplier theory” used to justify the 2009 stimulus. And he criticized the assumption that proposed spending equates with actual spending, which trickles out over time.

Aside from that, Mrs. Lincoln, the Goldman analysis was spot on.

‘Alchemists and Quacks’

This fundamental disagreement among professional economists about whether government spending helps or hurts represents the state of the art, or science, today. In what other science do practitioners design a treatment plan based on inconclusive proof that the medicine does any good?

There are no control studies in economics, no way to hold everything else constant to determine the impact of one variable, no way to falsify conclusions that models spit out. Financial Times columnist John Kay, writing yesterday about risk modelers, referred to them as “alchemists and quacks.”

A bit harsh, perhaps, but he’d probably hold macroeconomic models in the same high regard.

Whenever oil prices spike, modelers instantly project how much the increase will subtract from GDP growth. No mention of why prices are rising. Is it the result of a supply shock, which results in higher prices and reduced quantity demanded, or an outward shift in the demand curve, which equates with higher price and quantity demanded? There is a difference.

Known Knowns

In microeconomics, which is the study of how individuals and firms interact in specific markets, certain truths are self-evident. Which doesn’t mean economic planners can see them. Governments across Asia right now are using subsidies and price controls to ease the pain of higher oil and food prices even though their actions will exacerbate the crisis.

Goldman countered Taylor’s critique with a clarification. The projected 1.5 to 2 percentage point hit to GDP was to the quarterly annualized growth rate, not to the level. Thanks for that.

As I said before, we entered the 21st century with macroeconomics still looking for an Age of Enlightenment.

Five thousand years ago in ancient Egypt, medics used leeches to suck the blood of ill patients, believing the practice could cure everything from fevers to food poisoning.

Today’s physicians have largely forsaken bloodsuckers for modern medicine. It’s about time macroeconomics emerged from the Dark Ages as well.

Caroline Baum, author of “Just What I Said,” is a Bloomberg News columnist. The opinions expressed are her own.)
Dark Ages Indeed

I am wondering "How many times does an economic model have to be discredited before it is discarded?"

This idea that government spending can stimulate the economy is total nonsense. If it worked, we would see something more than 2.8% economic growth for a deficit of $1.4 trillion dollars.

The Fed purchasing Trillions of Fannie Mae and Freddie Mac bonds did nothing for housing, nor did several rounds of housing tax credits.

Government spending accounts for an ever-increasing share of GDP. Moreover, the only reason GDP is up at all is that by definition, government spending adds to GDP. The multiplier is actually negative. It takes an increasing amount of "stimulus" spending just to say in the same spot.

Taylor Model Nonsense

Taylor criticizes the Goldman multiplier model and rightfully so.

However, his own economic model is fatally flawed. He believes all the Fed needs to do is go on autopilot, hiking or lowering interest rates in accordance with the Taylor Rule.
In economics, a Taylor rule is a monetary-policy rule that stipulates how much the central bank would or should change the nominal interest rate in response to divergences of actual inflation rates from target inflation rates and of actual Gross Domestic Product (GDP) from potential GDP. It was first proposed by the U.S. economist John B. Taylor in 1993. The rule can be written as follows:
i_t = \pi_t + r_t^* + a_\pi  ( \pi_t - \pi_t^*  )  + a_y ( y_t - \bar y_t ).
In this equation, \,i_t\, is the target short-term nominal interest rate (e.g. the federal funds rate in the US), \,\pi_t\, is the rate of inflation as measured by the GDP deflator, \pi^*_t is the desired rate of inflation, r_t^* is the assumed equilibrium real interest rate, \,y_t\, is the logarithm of real GDP, and \bar y_t is the logarithm of potential output, as determined by a linear trend.
Unmeasurable Economic Gibberish

The idea that interest rates can be set by mathematical modeling when the variables themselves are subject to debate as to how to measure them is preposterous.

Take the CPI for example. I believe home prices should be in the CPI. They used to be.

Somewhere along the line some theorist decided "owners' equivalent rent" (OER) was a more valid concept. What is OER? It is the amount one would pay himself if renting a house from himself. It is the single largest component of the CPI. The measure of inflation from 2002 to now would be wildly different if one used actual home prices instead of OER.

Which model is more accurate? Look at the Fed's chasing-its-tail actions hiking in baby steps on the way up, then lowering interest rates to zero when the economy collapsed.

ECB President Jean-Claude Trichet, a Keynesian Clown Too

Just today, Jean-Claude Trichet is talking about hiking rates in Europe.

His concern is pass-through inflation as noted in the Bloomberg article Trichet Says ECB May Raise Rates, Show `Strong Vigilance'

“There is a strong need to avoid second-round effects,” Trichet said, calling for moderation from wage and price setters. The ECB is “prepared to act in a firm and timely manner.”

This whole idea of pass-through inflation and second-round effects is yet more Keynesian claptrap. If someone pays more for gasoline, they have less to spend on clothes. It is as simple as that, but not to those purposely hiding behind economic models and their multiplier effects.

Alchemists and Quacks Galore

Making decisions on flawed models is bad enough in closed economic society.

Errors in every model are exacerbated by the fact we have a global economy subject to economic pressures of all kinds from countless places.

Financial Times columnist John Kay, writing yesterday about risk modelers, referred to them as “alchemists and quacks.” There are no control studies in economics, no way to hold everything else constant to determine the impact of one variable, no way to falsify conclusions that models spit out.

On that basis, the analyst from Goldman Sachs, Taylor, Bernanke, Krugman, Greenspan (and countless others) are all quacks.

Why Model at All?

There are no control studies because it is impossible to do them.

The real world is constantly changing, while mathematical models, Goldman's and Taylor's alike sit there as unmeasurable economic gibberish, when every component is subject to measurement errors and debate about what needs to be measured in the first place.

End the Fed

The free market could not possibly have done a worse job in setting interest rates than the perpetual chasing-their-own-tail central bank tactics that continually create boom-bust bubbles of ever-increasing amplitude in both directions.

If central bankers knew where interest rates should be we would not be in this mess, or at least the mess would be smaller. For further discussion about what the Fed does and does not know, I strongly encourage you to read the Fed Uncertainty Principle.

Ironically, the one thing the Fed never mentions and the ECB seldom mentions is money supply.

Here's the deal: Inflation is a direct result of the cheapening of money. Strike that, inflation IS the cheapening of money and central bank policy in conjunction with fractional reserve lending is the cause.

Central bankers do not talk about such things because they are at the root of the problem.

The solution of course is to not only get rid of the Taylor rule, but to get rid of the Fed, the ECB, and central bankers around the globe.

Mike "Mish" Shedlock
http://globaleconomicanalysis.blogspot.com
Click Here To Scroll Thru My Recent Post List

Last 10 Posts


Copyright 2009 Mike Shedlock. All Rights Reserved.
View My Stats