Markets Are Neither Efficient Nor Irrational (Part 1)

 |  Includes: DIA, EWZ, GSC, GXC, JJM, QQQ, SPY
by: John Overstreet

As we know, there are known knowns. There are things we know we know. We also know there are known unknowns. That is to say, we know there are some things we do not know. But there are also unknown unknowns, the ones we don't know we don't know.

--Donald Rumsfeld

In a brief telephone interview shortly after the announcement that Lars Peter Hansen, Eugene Fama, and Robert Shiller were to be awarded the Nobel Prize "for their empirical analysis of asset prices," Shiller said that our understanding of those prices is "very much a work in progress….(NYSE:W)e've learned a lot about asset pricing, but there is a basic human element in it that is irreducible….(NYSE:T)he field of finance will never completely understand asset pricing movements." Hansen struck much the same note in his interview with "We're making a little bit of progress, I think there's a lot more to be done. I think the…modeling challenges have (BEEN) made clear." In effect, asset prices constitute a "known unknown." We more or less know what we don't know.

In this article, however, I am going to argue that if Shiller is right about asset prices being irrational, then by his standard, the prices at the other end of the spectrum, those of basic consumption goods, must be "irrational," too. But, if the entire spectrum of prices is irrational, one must wonder about the soundness of a discipline that cannot demonstrate its core principles, particularly its description of the pricing of consumption goods, and can come up with no other account of price deviations from orthodoxy than the exogenous powers of "speculation" and "emotion."

I will argue that history clearly demonstrates that not only are we, as yet, incapable of modeling complex, nebulous, new-fangled asset prices, we are incapable of modeling the prices of basic staples such as food and energy, and that the conclusion must be that our unquestioned "knowns" are almost wholly unknown. Moreover, I will argue that the separate and distinct markets for assets and goods "know" something that is otherwise quite unsuspected by us.

In other words, we have two "unknown knowns": a set of flawed but rational assumptions (namely how asset and goods prices should behave) and an economic dynamic that clearly behaves as if it knows something that economic theory is so far incapable of contemplating.

Are Asset Markets Efficient?

The first question we should take up is whether or not the notion of efficient markets is as applicable to financial assets as it is to goods and services, where price is understood to have an uncomplicated relationship with supply and demand.

The prices for financial assets like stocks and bonds or even items of ego-driven consumption such as a rare object of some sort (art, etc) do not seem to be subject to an equilibrium between supply and demand, as many have pointed out (I recommend George Cooper's vivid and highly readable description of these issues in The Origin of Financial Crises). When stock prices rise, as Shiller suggests in Irrational Exuberance, demand tends to rise rather than fall. The underlying value of the asset in question gets lost in feedback loops and unrealistic expectations. Emotion gets the better of us.

This dichotomy between financial, or speculative, assets on one side and consumption goods and services (and labor, according to the Philip's Curve), on the other, sounds very plausible. The price of tea in China is determined by the fundamentals of supply and demand, but stock prices are idiosyncratic in comparison, subject to the whims of sentiment. Stocks are not moved so much by fundamentals as by psychology.

But, if you try to engage a traditional-minded sort of investor or analyst on the question of "alternative assets" like commodities, the poles suddenly reverse. Stocks and bonds have revenue streams and generate profits and dividends. We have metrics by which to value them. We can make estimates about future growth and use this or that metric with these or those sets of variables and assumptions to dress those kinds of assets with risk-reward trajectories. But, commodities have none of these attributes. A bushel of wheat or a ton of tin doesn't yield anything. Goods have no clear metrics from the perspective of classical investing.

Does Fundamental Analysis Work With "Alternative Assets?"

In markets such as oil, the most vital of the traded natural resources and a nonrenewable one at that -- where one would suspect cold, brutal economic logic to reign without mercy -- the notion amongst sensible investors seems to be that, unlike yield-bearing assets, which can be subjected to rational and profitable analysis, speculation and irrational fear drives all before it. In the late stages of the oil boom in 2007-2008, spot crude prices were shooting up even though supply was clobbering demand. (Also, see PDF).

Or consider a longer term oil scenario: if one had known in 1980 that a) global GDP would grow by 80% in real terms for the next twenty years as the Iron Curtain fell and the economic potential of continent-sized, energy-hungry developing countries was unleashed, b) that consumer prices would rise by well over 200%, c) that the two largest economies would experience stock market booms and almost unprecedented degrees of economic confidence, d) that unemployment would fall to thirty-year lows, e) all while, in the heart of oil-production country, a decade-long war between a revolutionary theocracy and a megalomaniacal Arab nationalist not shy about using chemical weapons would be followed almost immediately by another couple of rounds of war, sanctions, and a vigorous Western military presence -- if one had known all that in 1980 -- would one have bet that this resource would be up or down by 2000? As it turned out, of course, oil prices were to drop by over 50% in nominal terms over those two decades. One would have expected oil to be at $200, not $10. And, this was the story to varying degrees among all commodities: long-term collapses in nominal prices throughout decades of strong economic growth.

It is very hard to make the supply/demand data fit the oil story of that period, just as it was with the oil shock that emerged in the midst of the financial crisis five years ago. In fact, it is hard to make supply and demand jive with any commodity prices, including the tea in Mombassa and Kolkata and, presumably, China, the oil in the Persian Gulf, or cocoa in Africa and Brazil.

The point is that, from the academic side, there is a debate about whether or not financial assets are more "speculative" than they are driven by values, while it is assumed that the prices for goods, such as soybeans, sugar, and aluminum are determined by fundamentals. Among practitioners, the problem is almost perfectly inverted: equities and bonds can be rationally analyzed under the terms of fundamental analysis, while basic commodities, as old as Adam, are "speculative."

There is not necessarily a contradiction here. After all, Shiller's notion that "irrational exuberance" in the stock market can be identified by historically extreme levels in Cyclically-Adjusted P/E ratios (CAPE) is merely an application of a portion of Benjamin Graham's Depression-era methodology for evaluating individual asset prices, a staple of fundamental analysis, to the broader stock market.

There is not necessarily a contradiction, but there is a tension. The contrast in attitudes towards the pricing of goods on the part of academics and investors is striking.

Are Goods Markets Efficient?

So, what about goods? Has anyone ever been able to demonstrate that their prices are driven by supply and demand? I do not know of any clear demonstration of this bedrock principle of markets and economics, apart from the dated "Phillip's Curve."

While academics have explored the relationship between supply, demand, and price in commodity markets before, it turns out that relationship is a lot more complicated than one might have assumed. The data must be tortured with all manner of statistical techniques before it will confess the supply-demand creed, but no orthodox version of that creed has been produced. A recent paper with a brief summary of the scholarly debate about whether supply or demand is the main driver in commodity prices and which particular mechanisms transport those forces through the markets can be found here. In the end, each of these arguments moves along similar, implicit lines, which I would express thusly: "Assuming that price is determined by supply and demand, how can we make the data conform to that assumption?"

The notion that prices for goods are driven by the fundamentals of supply and demand seems to be both as intuitively plausible and as unproven as is the notion that equity prices should have a demonstrable link to dividends. Of course, supply and demand must have some role to play in commodity prices, just as profits and dividends must have a role to play in equity prices. Water is more plentiful than beer, so (presumably) it is significantly cheaper despite being more essential, just as companies must generate profits in order for investors to want to purchase shares. But, are those the decisive valuation elements of the respective prices of goods and assets above a certain threshold?


And, if they are not, what are the alternatives? If everything we know about valuations and basic economic principles is wrong, is there nothing to do but go 20% stocks, 20% bonds, 20% cash, 20% gold, and 20% in canned goods and ammo? That's not the worst idea ever (although one might want to lighten up on the ammo), but I believe we can do better than that.

Having argued that neither asset markets nor goods markets seem to be efficient in the narrow sense of the word, that is, insofar as they fail to conform to the assumptions of how those categories "should" behave, in part two, I will argue that it is extremely unlikely that either of these markets are irrational in any meaningful sense.

In fact, the valuations for stocks and goods are effectively identical. As I have contended in previous articles, the structural, secular, and cyclical narratives of the current market are all bullish for US equities (SPY, DIA, QQQ) and bearish for commodities (GSC, JJM), as well as for developing markets such as the BRICs (EWZ, GXC). In the post-gold standard world, stocks and commodities are mirror images of each other because of deep, chaotic forces and a human one, as well.

SPY vs China, commodities, and treasury yieldClick to enlarge

Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.