A low volume, high quality source from the demand side perspective.The podcast is produced weekly. A transcript is posted on the day of.

Thursday, September 16, 2010

Transcript: 404 The Four Major Forecasting Errors

Listen to this episode

Today on the podcast, we have a full boat, Hyman Minsky, presidential economics, the four main errors of economic forecasters and --- yes -- Idiot of the Week.

The last portion today is an extended excerpt from the Obama press conference from last week as it regards the economy, to give you a notion of what the political terrain looks like. And we -- like Robert Shiller -- are wondering why we don't reduce unemployment by hiring people. We offer a program from Hyman Minsky, proposing an economic stability policy based on employment rather than interest rates and corporate tax cuts. But first ... you may remember last January we spoke with the chief economist of the regional forecasting council and recommended he go with our bouncing along the bottom with chance of downside crises forecast -- the forecast we've had for a very long time -- but we could understand if he needed to keep closer to the consensus for personal professional reasons. Turns out he chose the latter, and in April issued an appropriately guarded but inappropriately optimistic forecast along with a call for higher inflation. He had to walk that back in July, which he did with aplomb, but he should have gone with the Demand Side stock. In any event, we are inspired to write a memo outlining the four basic forecasting errors that cloud the current outlook for the non-Keynesian, non-Demand Side economist. Which includes virtually all, and excludes only those who have been right.


  1. Assuming consensus is a good indicator
  2. Assuming the Fed controls the money supply
  3. Assuming a natural return to equilibrium
  4. Assuming a return of the consumer economy.

On the first he went so far as to statistically distribute and analyze the predictions of forecasters as if they were independent data points. Any empirical examination will show economists' forecasts more than a few quarters out to be highly inaccurate. Witness the major failure to predict the GFC and before that the recession brought on by the housing bubble and bust. Since most forecasters share the same Neoclassical errors, there should be no surprise that they correlate much better to each other than they do to actual outcomes. This is nearly explicit in a statement you often hear that forecasts are subject to a high degree of uncertainty.

On the second, the error of thinking the Fed controls the money supply is admittedly nearly ubiquitous, in spite of some detailed work done by no less conservative economists than Edmund Phelps and Finn Kydland. We've made this point before, but it bears repeating. The money supply expands and contracts to fit the boom and bust. That means innovation and intermediation on the way up and deleveraging and disintermediation on the way down. These processes are the V in MV equals PQ. When economists say we are sure to have inflation because the Fed is creating money, they are wrong. The Fed can make base money, but it cannot make the banks lend, nor people borrow, nor the rest of the story.

As John Kenneth Galbraith once noted, nothing has survived contradictory evidence so well as the belief in the efficacy of monetary policy. The Fed has been able to stall the economy quite effectively at times by increasing interest rates and the cost of doing business, but in terms of starting it, not so much. This follows directly from the mistaken belief that the Fed controls the money supply and the money supply will expand in proportion to the Fed's printing.

The third mistake is also endemic to Neoclassical economists. That is the widespread belief that there is a natural stabilizing mechanism of one kind or another. Sometimes it is postulated through the labor markets, more recently you hear that corporate profits will stimulate investment. As we noted before, these profits were generated by cutting costs -- efficiencies they are called. This means jobs and this is downward pressure on the economy as a whole, even as it pushes up company bottom lines. No matter the speculation on labor right-sizing its wage demands, no market function exists that brings economies into the mythical equilibrium naturally.

So-called automatic stabilizers do exist, but these are the policies and programs of social insurance -- unemployment and social security -- of the New Deal that kick in automatically. They are not anything inherent to capitalism, but an adjustment to capitalism. These put a floor under demand, and often by increasing government deficits. Discretionary -- as opposed to automatic -- stabilizers are those things like public works or jobs programs that are passed during the crisis. They are no less functions of public policy action, they are just not embedded in law prior to the need for them. These are the base under demand, and in the current crisis, these are the reasons the depression has not been as deep as that of the 1930s, where they were invented in the New Deal. Federal revenue sharing today with states and local governments would be an extremely effective stabilizer. But it difficult to institute politically for the opposition of the savage stupidity generated by panicking self-interest.

Fourth, and last on this list of major forecasting errors, is the nearly universal assumption on both Left and Right that any recovery will come through the private consumer economy. This ignores the huge private debt that crushes spending power, but also the general absence of productive investment opportunities in consumer products. A glut of capacity exists. And right in the middle is the housing situation. Housing has led every recovery since the Second World War. Housing was nurtured by growth-oriented economists since the great Leon Keyserling, for its ability to produce both jobs and household assets. Now there is a tremendous overhang of inventory and the housing asset is now a financial liability.

Were demand to be shifted to public goods, such as infrastructure, energy and social services, ample opportunity for private investors would be forthcoming. It is investment that produces growth in both the short and long terms. It is the prospect of profit that stimulates investment. It is more or less certain demand for products that produces the prospect of profit. This is the reason for advertising, to control demand for the goods one wants to produce. But even with modern mind control, the consumer is not a source of stable, certain demand, because he is tapped out.

The government can be.

And before we leave this, we have to note again, that economics is being put to the service of the status quo, and the status quo is unsustainable, not least in its utilization of the environment. We are at a tipping point with regard to the planet's climate. This tragic fact is ignored in the economic debate when it ought to be at the center. Next week, we'll look at energy and carbon as a route out of the economic mess. Today, lets just recognize that reorganizing and retooling and reconstructing transportation, housing and energy will pay off in real terms, and we will be crippled in all dimensions until we can see this.

Now, on the subject also to be covered later in the excerpt from the president's press conference, we have Michelle Gerard of RBS



Stop, Stop Stop. 97 percent of small businesses are not impacted. Of the three percent that are, some are in the Fortune 100, not so small, but these are S corporations. Beyond this, when people say small business creates the most new jobs, they are primarily talking about new businesses which succeed. That is, the Whole Foods which go from one store to one thousand and explode into big business. These companies are not going to worry about less than five percent on the top marginal income tax. Another group comes from lawyers and doctors and other high-priced professionals who are small businesses covering a group of professionals.

Is this a headwind? We illustrated the effect of tax cuts on the rich as a means of stimulating the economy a few weeks ago using the multiplier. It turns out that high-roller tax cuts have a multiplier below .6, as empirically derived. This means for every one dollar of benefit, they produce sixty cents of economic action. This is a loss of 40 cents.

And finally, if there were any sound economic theory behind this, you can be sure it would be trotted out right now. Instead we get the weather report, headwinds, or logical fallacies: because it works for the middle class, it has to work for the rich. No, it doesn't. This is not class warfare. It is legitimate economics.

Michelle Gerard


Idiot of the Week.

Robert Shiller recommended a couple of weeks ago an employment program to stabilize the economy. Figuring $30 billion per one million jobs. At that rate, for the cost of one AIG bailout, we could have three percent unemployment. But Hyman Minsky was far ahead. Still, of course, following the great John Maynard Keynes and the masters of the New Deal, but far ahead of today.

While current economic stability planning and policy has been ceded to the Fed, with its short-term interest rate, and to inducements to corporations, Minsky saw twenty-five years ago that a better approach than interest rates was employment.


Stabilizing an Unstable Economy, p. 343

An Employment Strategy

Although stabilization policy operates upon profits, the humane objective of stabilization policy is to achieve a close approximation to full employment. The guarantee of particular jobs is not an aim of policy; just as with profits, the aggregate -- not the particular -- is the objective.

The current strategy seeks to achieve full employment by way of subsidizing demand. The instruments are financing conditions, fiscal inducements to invest, government contracts, transfer payments and taxes. This policy strategy now leads to chronic inflation and periodic investment booms that culminate in financial crises and serious instability. The policy problem is to develop a strategy for full employment that does not lead to instability, inflation and unemployment.

The main instrument of such a policy is the creation of an infinitely elastic demand for labor at a floor or minimum wage that does not depend upon long-and short-run profit expectations of business. Since only government can divorce the offering of employment from the profitability of hiring workers, the infinitely elastic demand for labor must be created by government.

A government employment policy strategy should be designed to yield outputs that advance well-being, even though the outputs may not be readily marketable. Because the employment programs are to be permanent, operating at a base level during good times and expanding during recession, the tasks to be performed will require continuous review and development.

There are four labor-market aspects to an employment strategy:
  1. The development of public, private and in-between institutions that furnish jobs at a noninflationary base wage.
  2. The modification of the structure of transfer payments.
  3. The removal of barriers to labor force participation
  4. The introduction of measures that constrain money wages and labor costs.

The four aspects of the employment strategy are linked. If the massive transfer-payment apparatus is to be dismantled, then alternative sources of income must be guaranteed for the current and potential recipients of such payments. If barriers to labor-force participation are removed, then jobs have to be available for those who are now free to enter the labor market. Constraints upon money wages and labor costs are corollaries of the commitment to maintain full employment.

Now and finally, an extended excerpt from last week's presidential news conference. Already the ice seems to be breaking as a result of this no doubt political initiative. But here, without further introduction:



  1. Although stabilization policy operates upon profits, the humane objective of stabilization policy is to achieve a close approximation to full employment. The guarantee of particular jobs is not an aim of policy; just as with profits, the aggregate -- not the particular -- is the objective.

  2. Robert Shiller recommended a couple of weeks ago an employment program to stabilize the economy. Figuring $30 billion per one million jobs.