A Clean Kill of the Keynesian Demand Shortfall Error

This excellent piece on Say's Law makes a powerful point in plain language: the only way aggregate demand can exceed aggregate supply is if new credit is created, thereby enabling spending before income or consumption before production. In the short-run, the Keynesian  rebuke to Say's law works because balance sheets of households, businesses and the state can be leveraged-up on a one time basis, and Keynesian policy has been doing essentially that since the 1930s---but especially since the reign of monetary central planning 25 years ago.

The trouble is that on an intermediate basis the monetary stimulus to credit expansion can be over-done, and we get the boom-bust cycle that central banks have made the patented feature of their baleful craft. But the central banks can only ignite the boom-bust cycle so many times because it is linear, not circular and timeless. We do not go back to "go" and restart. Instead, with each boom-bust cycle we use up more of our balance sheets to borrow and spend without first producing. At length, we hit "peak debt", which is where the US economy, Europe, Japan and even China are today.

At the point, the Keynesian witch doctors, like Prof. Larry Summers, sputter that the state's fiscal and monetary branches are not "stimulating" enough and that we therefore need more in order to attain "escape velocity".

Well, no. In fact, the arc of official GDP, which purports to measure aggregate demand, is simply being pulled down to the level of aggregate production and income. There is no balance sheet left by which artificial current demand can be stolen from future production. The Keynesian Interregnum is over. Say's Law is back!

Say’s Law and the Permanent Recession

Mises Daily: Friday, February 28, 2014


Mainstream media discussion of the macro economic picture goes something like this: “When there is a recession, the Fed should stimulate. We know from history the recovery comes about 12-18 months after stimulus. We stimulated, we printed a lot of money, we waited 18 months. So the economy ipso facto has recovered. Or it’s just about to recover, any time now.”

But to quote the comedian Richard Pryor, “Who ya gonna believe? Me or your lying eyes?” A Martian economist arriving on earth would have to admit the following: the US economy has experienced zero real growth since 2000. This is what I call the permanent recession. Permanent, because, unlike past downturns — there will be no recovery. To make the case for this view, I will rely on the ideas of several classical and Austrian economists: J.B. Say, James Mill, Mises, Rothbard, W.H. Hutt and Robert Higgs.

I will begin with the J.B. Say, who is known for the eponymous Say’s Law. To explain I will quote from Say’s Treatise on Political Economy:

[A] product is no sooner created than it, from that instant, affords a market for other products to the full extent of its own value. ... Thus the mere circumstance of creation of one product immediately opens a vent for other products.

Say’s Law can be explained in the following terms:

1. The way that a buyer demands a good is by supplying a different good.

2. The supply of one type of good constitutes the demand for other, different goods.

3. The source of demand is production, not money. Money is only a temporary parking place for past production.

In the modern economy with division of labor, most of us demand goods when we supply our labor. I work as a software engineer. I supply my labor writing computer software. And from that supply I am able to demand other goods, such as coffee.

Say’s ideas were used to settle a debate between the British economists David Ricardo and Thomas Malthus who believed recessions were caused by a general glut. The concept of a glut for a single good is easy enough to understand: there is more supply on the market than demand at the offered price. A glut can be alleviated by a fall in the price of that good. The producers of the good may take a loss if the market price is below their costs, but the market can always clear at some price.

The idea of a general glut is that all markets for all goods are in surplus. And for some reason, prices are unable to fix the problem. Ricardo opposed Malthus, arguing that the concept of general glut violates sound economics and clear thinking. He argued this point using Say’s Law: because demand is constituted by supply, aggregate demand, meaning the demand for all goods on the market, consists exactly of all things supplied. Aggregate demand is not only equal to, but identical to, aggregate supply. The two can never be out of balance. And if a general glut is a logical impossibility, then it cannot be the cause of a recession.

The idea of aggregate supply and demand in getting out of balance has appeared many times in the history of economic thought. The same idea is either called overproduction or underconsumption, depending on whether the problem is too many goods or not enough purchasing power. Keynesian economics is a form of underconsumption theory. The overproduction/underconsumption theory has been debunked by sound economists, but like a zombie, it refuses to die.

It is acknowledged by both sides that, if Say’s Law is true, then Keynes’s entire system is wrong. Keynes knew this, so he took upon himself the task of refuting Say’s Law as the very first thing in his General Theory. Keynes’s argument was that Say’s Law is only valid under the conditions of full employment, but that it does not hold when there are unemployed resources; in that case we are in the Keynesian Zone where the laws of economics are turned upside down.

But, as Stephen Kates explains in his book Say’s Law and the Keynesian Revolution (subtitled How Economics Lost its Way), Keynes failed in his attempt to overturn Say’s Law. Kates shows beyond any dispute that Say and his fellow classical economists were well aware that there could be unemployed resources, and that Say’s Law was still valid in that case.

To summarize, there is no such thing as a general glut or a demand deficiency, we can have idle resources, and Say’s Law is still valid. So how did classical economists explain recessions? Producer error. Producers had produced the wrong mix of goods. James Mill in his essay “Commerce Defended” explains the meaning of producer error:

What indeed is meant by a commodity’s exceeding the market? Is it not that there is a portion of it for which there is nothing that can be had in exchange. But of those other things then the proportion is too small. A part of the means of production which had been applied to the preparation of this superabundant commodity should have been applied to the preparation of those other commodities till the balance between them had been established.

Kates and Gerard Jackson have argued that the classical economist had a theory of producer error much like the one later developed by Mises. Mises developed existing ideas and integrated Austrian capital theory and time preference theory to provide an explanation of why many producer errors occur at the same time. We know this as the Austrian theory of the business cycle.

Mises called the production errors malinvestment. These errors happen systemically because of fractional reserve banks loan money into existence that is not backed by savings. That misleads producers into thinking that there are more real savings available than society wishes to save. Producers then make both the wrong mix of capital goods of different orders, and the wrong proportion of capital goods in relation to consumption goods.

When there is malinvestment there must be a recession, for the following reason: there were never enough real resources to complete all of the capital projects that were started during the boom. The firms that started these projects either over-estimated the demand for their output, or, under-estimated their costs. Somewhere along the way, firms will discover that they cannot obtain all of the factors they need at a price below their costs. They cannot make profits. Many of them fail.

I will give an example of how malinvestment leads to a recession. I worked in San Francisco during the tech bubble. There were many tech startups. Each one assumed that they would be able to grow by hiring more employees at the prevailing wage rates. But the prevailing wages did not reflect the true scarcity of skilled technical people because all of these businesses planned to hire more workers over the same time frame. But the number of skilled engineers could not possibly grow that fast. If you think of a software engineer as a form of human capital, a software engineer has a long period of production and requires many inputs (mostly coffee, but some other things as well).

And there just weren’t enough engineers to build all of these web sites. One firm could have hired more engineers by paying double the prevailing wage, but it wasn’t in the economics of their business to do so. And in any case, most of the compensation was in imputed value of the stock options, which could be any number that you want if you assumed that the bubble will keep blowing up forever.

What this shows is that, while you can fund a new venture with money printing, you cannot print skilled workers or office space. At some point, real factors become the bottleneck, so their price has to rise. And when that happens, some producers get squeezed out because they cannot raise prices. If they over-estimated demand for their output from the start, they would have needed lower, not higher, costs to make profits. And that is the start of the recession.

Mises’s theory explains why the boom starts and why it comes to an end. Production errors cannot continue indefinitely because they result in losses. But why do we have a lasting recession? Why doesn’t everyone find a new job tomorrow? To explain, I will turn to the great English Austrian, W.H. Hutt.

Hutt used Say’s Law to explain the recession. Hutt observed that when one person becomes unemployed, he stops producing — and supplying. And from Say’s Law he loses his power to demand. Also from Say’s Law, his demand constituted the means for others to supply, and for those others to demand, so the prosperity of others is diminished. The malinvestments are the first ledge in the waterfall. Then, other businesses will see the impact, even those that were not originally part of the malinvestment.

Keyes said something like this in his model of the circular flow of spending. Keynes was right that there is an interdependence of all economic activity. But Keynes was wrong about consumption being the driving force of this: it is producing, not consuming. According to Say, the interdependence is constituted by the relationship of all production, not of expenditure. Expenditure of money is only the culmination of the process that began with production:

That each individual is interested in the general prosperity of all, and that the success of one branch of industry promotes that of all the others. In fact, whatever profession or line of business a man may devote himself to, he is the better paid and the more readily finds employment, in proportion as he sees others thriving equally around him.

I will here give another example. When I worked in San Francisco during the tech bubble, I got coffee every day at a café near my office. When the tech bubble burst, this café failed as well. Was that because they made bad coffee? Or because software engineers got tired of drinking coffee? I can assure you that did not happen. It was because customers at the café were part of the bubble. The difference between pets.com and the café is that, had there not been a bubble, the same café would have existed at the same spot, serving coffee to different people working at different jobs producing different things that were demanded by a balanced market, because people who go to work still want their coffee.

Hutt also had an economic explanation of how the economy recovers from a recession. He emphasized that any useful good or service can be integrated into the price system, somewhere, and at some price. Once someone is again producing, he can supply, and then he can demand, and by demanding, he creates a market for the supply of other producers. And so on. But that requires two things: time and flexibility.

It takes time for entrepreneurs to sort through the broken shards of the boom to figure out what is really in demand, and what the supplies of factors are. But the recovery will occur because eventually entrepreneurs see all of those unemployed resources as a bargain. Productive assets and labor won’t stay on sale forever. When prices of some factors get low enough, then the people who held on to some cash will see attractive yields.

Either people will move around, or just take the best job that they can in order to get by until things improve. The empty offices will get leased out. The key is that profit margins must open up. Hutt argued that can happen even in a depression if prices are flexible because there is always some way to combine inputs into outputs at a profit, if prices will cooperate. When confidence is low, entrepreneurs will make more conservative estimates of the market for their outputs, and they may require wider profit margins. You can think of it as a risk premium. And that means that the prices for some types of labor and capital must fall considerably not only from their bubble values, but in relation to other prices.

During the tech bubble, many small companies were formed. Every one of them required a director of engineering, a CEO, a CFO, and several other key management positions. People got hired into these jobs who lacked the experience to get such a job in an established company, or people at established companies were hired away and less experienced people were promoted. This process could be described as job title inflation: high level job titles were debased. When the recession hit, a lot of these positions simply went away, and the people who held them had to seek new jobs. Some of these people rose to the level of their new responsibilities and advanced their career, but others had to take a step back and accept a lower paying job with a less important-sounding title. And if they were unwilling to do so, then that prolonged the duration of their unemployment.

When there is no recovery, or it is long in coming, what got in the way? Hutt had an answer to this: the price system is blocked from working. Hutt emphasized wage rigidities caused by labor unions. Unions are cartels that attempt to create a monopolistic price by legally raising wages above labor’s marginal value product. When the price of something increases we move along the demand curve to a lower quantity demanded. Less quantity means less labor is hired into those jobs, and the displaced workers must find some other work, which is by definition lower paid or otherwise less preferable.

Hutt explained why labor unions decrease aggregate demand rather than increase it. When workers shift from a higher-valued occupation to a lower-valued one, they produce less, and therefore supply less, and by Say’s Law, demand less. And as Hutt showed, by demanding less, they diminish the market for the supply of others.

Anything that prevents wages or asset prices or capital market prices from falling moves markets away from clearing. In the modern world, one of the main barriers to recovery is Keynesian stimulus. Stimulus tries to create more demand without creating more supply. We know from Say’s Law that this is doomed to fail because supply and only supply constitutes the demand for other goods. What stimulus is really trying to do is to inflate the fake price system of the boom so that more expenditures can occur at the fake prices producing more of the wrong things for which there was never a real demand in the first place. And that cannot work because it was the breakdown of production under the fake prices that caused the boom to end. For a real recovery to occur, production must be reorganized along the lines of consumer demand.

Now I am going to turn to the Great Depression and show the relevance of Hutt’s thinking to that time.

Prior to the 1930s, recoveries from a panic, as they were called, took about 12-18 months. The great depression of 1920 (the one that you’ve probably never heard of) lasted about that long. Why? As historian Thomas Woods wrote, “Harding cut the government’s budget nearly in half between 1920 and 1922. The rest of Harding’s approach was equally laissez-faire. Tax rates were slashed for all income groups. The national debt was reduced by one-third.”

The Great Depression also began with a stock market crash followed by a downturn. As the price system began to work, a normal recovery had begun by the early 30s. Then the New Deal kicked in, which created a depression within a depression that lasted until the mid-1940s. Ten years later, what could have kept the US economy underwater for 15 years? The price system was blocked, especially in labor markets.

Herbert Hoover held to a variation of underconsumption theory called the purchasing power theory of wages[1] According to this theory, high wages in themselves created more purchasing power. And by “high” he meant, above market values. Low wages, thought Hoover, were the cause of the depression because labor did not have enough purchasing power to buy back its own output. Hoover exhorted business leaders not to lower wages, and many of them believed him and followed his advice, or did so because they were clear enough that regulation would follow had they not complied. As explained by standard price theory, Hoover’s policies produced unemployment on a massive scale.

Hoover also believed in a strange class warfare doctrine. He thought that by preventing wages from falling, that all of the burden of the adjustment of production could be shifted from labor as a class to capital as a class. In America’s Great Depression, Rothbard quotes Hoover as follows:

For the first time in the history of depression, dividends, profits, and the cost of living have been reduced before wages have suffered. ... They have maintained until the cost of living had decreased and the profits had practically vanished. They are now the highest real wages in the world.

Following Hoover was FDR, who made things even worse. One of the New Deal agencies, the National Recovery Administration, employed agents to scour the country looking for stores that were lowering their prices. From Jim Powell’s FDR’s Folly: How Roosevelt and his New Deal Prolonged the Great Depression:

There were some 1,400 NRA compliance enforcers at fifty-four state and branch offices. They were empowered to recommend fines up to $500 and imprisonment for up to six months for each violation. On December 11, 1933, for instance, the NRA launched its biggest crackdown summoning about 150 dry cleaners to Washington for alleged discounting.

In addition to problems with prices, there was a deeper problem with the New Deal. For that, I will turn to the contemporary economic historian Robert Higgs.

Higgs has noted that the Great Depression was characterized by a collapse in capital spending. Austrians know that capital accumulation is what increases real wages. And capital consumption means lower real wages. We also know that a large volume of gross capital investment is required to offset capital simply wearing out from use every year. Net capital investment begins only when gross investment more than offsets capital consumption. And there is nothing to ensure that any volume of gross investment at all must occur during any given year. If investors stop investing, then the capital stock shrinks, and real wages, even under conditions of full employment will fall.

The reason for the collapse in investment was, says Higgs, “a pervasive uncertainty among investors about the security of their property rights in their capital and its prospective returns.” Higgs calls this regime uncertainty. This rational fear was based on the ideology of the New Dealers. The New Deal brain trust was full of anti-market ideologues who wanted to restructure the US economy from a free enterprise system to a socialistic-fascistic centrally planned system.

Higgs gives several pieces of evidence in support of the regime uncertainty hypothesis. One, qualitative, was the writings of business leaders in which they explained their reasons for lacking the confidence to invest. Second, opinion polls showing the same thing. And the third was the sharp rise in the term premium of corporate bond yields at maturities beyond one year. While we can only guess at the reasons for this, Higgs points out that it was not the usual yield curve that we are accustomed to in bond markets. Higgs attributes this heightened risk premium to the extreme levels of uncertainty investors had about the future of property rights in equity and debt, which are long time horizon assets.

And now I am at the point that I promised in the title. It is my view that we have been in a recession since 2000, that the economy has not recovered, and will not recover. I will first provide some supporting evidence that the economy is in a recession, and then explain why.

John Williams, an economic statistician and the proprietor of the web site Shadowstats, has produced a version of the real GDP based on the government’s nominal GDP deflated by his own GDP deflator. (The GDP deflator is sort of like the CPI, a price index that is used to convert nominal GDP into real GDP. For some reason they don’t use the same price index for both consumer prices and for this.) Like Williams’s own CPI, his GDP deflator is computed with older rules from before the time when the BLS began cooking the books to hide inflation. Williams’s measure of real GDP shows low to negative growth over the period since 2000.

Another way of measuring the economy is through the capital stock. The US economy requires about a trillion dollars per year of gross investment just to replace capital consumption. Higgs has written that real net private investment for 2012 was at an indexed level of 60 compared to a baseline of 100 for 2007. Corporate America is sitting on huge piles of cash rather than investing it. American non-financial corporations hold more cash than they have for 50 years.

Many measures of labor markets show zero to negative wage growth. While this is to some extent due to problems in labor markets themselves, a shrinking capital stock should show up as lower wages. Look at the labor force participation rate: it is now at the lowest level in decades and is plummeting rapidly. Also check out the trend in median household income. Analyst Jeff Peshut at RealForecasts publishes some similar graphs showing the negative trends in the volume of employment in labor markets.

Anecdotally, the media frequently reports that new college graduates cannot find career path entry level jobs, so they are forced to enter the labor force on a low wage track doing relatively unskilled work.

Another way of looking at the size of the economy is through the rate of time preference. Higher time preference means less saving, less investment, and less capital accumulation. But how do we measure it? Interest rates and yields generally of all kinds of assets, both financial and corporate balance sheets, are a measure of this.

Profit margins reflecting internal yields on US corporate assets have increased in the last few years. According to what Andrew Smithers disparagingly refers to as “stock broker economics,” high rates of profit are good for stocks. The Austrian economist Jesús Huerta de Soto makes an under-appreciated point about profit margins and stock prices.[2] Pervasively high or increasing rates of profit may show that the rate of time preference is increasing, implying that the capital stock is shrinking. If not time preference, then the perception of risk may be increasing, which would have a similar depressing effect on investment.

Given the work of Hutt and Higgs in explaining why a recession persists with no recovery, here is a list of factors causing price inflexibility and regime uncertain in today’s economy:

1) Capital market price floors, like the Greenspan-Bernanke put and QE which prevent the markets for capital goods from clearing.

2) Bailouts of Wall Street, which are another form of price floors, and keep the incompetent management teams in place.

3) The nationalization of the mortgage market, another form of capital market price floors and house price floors, which removes the largest sector of credit markets from the domain of economic calculation.

4) Obamacare. Besides the direct costs for taxpayers, the bill introduces massive incentive changes in labor markets, the implications of which are still not clear.

5) Economist Casey Mulligan documents extensive changes in labor market incentives in his book The Redistribution Recession. He argues that these changes have created a huge implicit tax on income for the unemployed contemplating an offer of paid work.

6) The pending default of most pension plans including Social Security, the medical welfare state, US states, counties, and cities. How the default will be paid for is creating great uncertainty.

7) Uncertainty created by the threat of wealth taxation and bail-ins, as outlined in an IMF paper.

8) The surveillance of all financial transactions and expanded reporting requirements for the assets of wealthy investors

As Hayek said, the more the state centrally plans, the more difficult it becomes for the individual to plan. Economic growth is not something that just happens. It requires saving. It requires investment and capital accumulation. And it requires the real market process. It is not a delicate flower but it requires some degree of legal stability and property rights. And when you get in the way of these things, the capital accumulation stops and the economy stagnates.

David Stockman's Contra Corner is the only place where mainstream delusions and cant about the Warfare State, the Bailout State, Bubble Finance and Beltway Banditry are ripped, refuted and rebuked. Subscribe now to receive David Stockman’s latest posts by email each day as well as his model portfolio, Lee Adler’s Daily Data Dive and David’s personally curated insights and analysis from leading contrarian thinkers.

Get Access