Central bankers such as those of the European Union and China have accepted the blame for the recent ballooning of their own real estate markets and have taken measures to slow it down. But the American Fed is loathe to do so “without further examination of the data.”
It is no surprise that in self defense they have published research on the subject, one example of which appeared in a paper in the Chicago Fed’s Economic Perspectives called “The great turn-of-the-century housing boom.”
This research is authored by in-house economists Jonas D.M. Fisher and Saad Quayyum, and concludes: “[I]t appears that the housing boom has not been driven by unusually loose monetary policy.”
The authors do an excellent job of identifying four potential sources of the housing price explosion. They are:
The Fed’s loose monetary policy. This fits the classic pattern of booms, and seems to fit the current case.
But the Fed’s econometricians have come up with three more market fundamentals as possible sources:
An increase in consumer real wealth, brought about by dramatic technological progress over the previous decade. This would encompass the increased productivity achieved through a more successful implementation of the internet, robotics, communications improvements, and other great technological advances over the years.
We could call this a case of real dollars liberated by the technology sector, chasing after a heretofore stable supply of goods in the housing sector. No one can deny that such progress has been made; but the question that jumps off the page is: Why do those liberated real dollars alight in the real estate market cause price increases and supply increases but not affect all sectors in this way? This question the Fed economists think they have answered with numbers 3 and 4, below.
Next, they suggest that changes in the demographic, income, educational, and regional structure of the population could be a factor. This would imply that the labor pool is better educated, that they are moving around the country more thereby putting pressure on particular real estate markets like Los Angeles or Las Vegas, plus that they have an increased home-buying desire due to their increased earning capacity.
Here we would have a supposed increase in the number of real dollars through an increasingly educated and technically savvy workforce, and these people would be physically more mobile and financially more wealthy and able to purchase homes. Thus, increased real dollars are again chasing after a stable supply of goods.
Unfortunately, the authors use the increase of net wealth during the 1990s as a confirmation of this hypothesis; but “net worth” is a function of the elevated nominal valuations of homes, and therefore increased net wealth may also not be real.
The fourth potential market fundamental they find to explain the housing price explosion is technology-driven development specific to the mortgage market that has led to an increase of more secure subprime lending and risk securitization. This is a reference to the changes taking place in the mortgage market (longer terms, more adjustable rates, interest only, no down payment, etc.), and the changes also taking place in the borrowing parameters (adoption oflooser credit requirements), both of which have opened the doors to a larger borrowing pool.
This would explain the redirection of investment dollars towards the housing market, or what we could call more confident dollars chasing after too few real estate goods.
In spite of the fact that these hypotheses do not address the underlying question of the origin of the dollars that are being lent, these descriptions of real estate market fundamentals will most likely meet with general approval among a majority of economists. Most would agree that any of these factors could indeed contribute, alone or in concert, to the observed price increases.
From this point on, however, the article’s authors are on shaky ground, even from their own methodological perspective.
The Fed, along with many in the economic community today, like to think of themselves as empiricists in the truest sense of the word, so it is interesting to examine their presentation from the point of view of their own brand of science.
What is the foremost criterion for sound science among the Fed’s own econometrician contemporaries? The answer should be “peer review,” and for a standard, we can refer to the Office of Management and Budget’s December 15, 2004 Final Information Quality Bulletin for Peer Review, which defines it as “an exchange of judgments about the appropriateness of methods and the strength of the author’s inferences. Peer review involves the review of a draft product for quality by specialists in the field who were not involved in producing the draft…. The selection of participants in a peer review is based on expertise, with due consideration of independence and conflict of interest.”
Given the Fed-scientist paper’s venue, we know that they did not publish it with the intent to invite examination and criticism of their methodology from economists outside their own school, nor even from economists within it, for that matter. This is something they rarely do, if ever, and Economic Perspectives is a Fed-published journal, not a purportedly independent arbiter-publisher such as the American Economic Review, the Journal of Monetary Economy, Contemporary Economic Policy, or the Quarterly Journal of Austrian Economics, where economists can exchange ideas and critique each other’s drafts.
Without submitting their work to preliminary review, they are, according to their own precepts, opening themselves up to an accusation of conflict of interest, and to seeing their valiant defense effort deflated down to a mere decree on the subject by the powers that be.
Mysteriously, however, their economics colleagues are remarkably silent, for reasons probably having to do with the Fed’s political and professional prestige. The Fed is not held to any particular standard, in spite of the fact that they should be all the more cautious because they do hold inordinate responsibility for our money supply.
The second essential element of good science according to the scientific method is the quality of the tools utilized. In medicine, for example, one tool is the double-blind study, in which technicians administer new medications or placebos to selected patients under a coded system whereby neither the administrator nor the patients know the nature of the medication, or which patient is getting what. This double insurance of impartiality affords accurate results as a rule, but even with as good a track record at it has, the double-blind study does not guarantee perfection.
Researchers must do further testing, and they must maintain their vigilance even once the product has been released for public consumption. The recent case involving Vioxx is an illustration of the hazards involved. Like the human body, each evolving economic situation has the potential to be unique, even though it may resemble others.
As noted frequently by the economists of the Austrian School, utilization of such a controlled observation technique is impossible in the field of economics. Economists cannot subject an entire economy or any element of it to such managed scrutiny. They have no substitute laboratory animals or volunteer study groups at their disposal; and so use of empirical methods of research does not bear perfect fruit.
In its stead, researchers who hold onto their faith in empiricism have developed a technique called econometrics, which is based on the use of mathematical models and statistics. Econometrics attributes symbols to certain elements in a dynamic economy as though they were (i) identifiable and (ii) static, and then the econometricians manipulate these symbols, and the statistics pertinent to them, in ways that can sometimes help them understand the different relationships among the elements they represent, at least in theory.
Econometricians claim that they have had some success with this methodology. However, even they admit that its usefulness depends upon the viability of multiple assumptions, differentiations, and fixed relationships that don’t necessarily exist in real life; and the testing of any conclusions is limited to the controlled context of the blackboard. Any and all hypotheses must be continuously reexamined over time; but economic problems need immediate solutions, so some researchers are tempted to jump to conclusions rather than wait for confirmation.
The paper itself describes the authors’ awareness of the precariousness of econometrics as a tool. For example, they state their main hypothesis thus:
This article does not address home prices directly. Rather, it seeks to understand recent developments by focusing on quantities. To the extent that the quantities can be understood by considering the underlying economic fundamentals such as productivity growth and the evolution of the mortgage market, then the recent growth in house prices is probably not due to excessive speculation in the housing market, such as occurs in a bubble. We argue that our findings point toward the high prices being driven by fundamentals.
The use of phrases like “to the extent that the quantities can be understood,” “it appears that,” “seem to,” “if true,” “are likely to,” “given the paucity of data,” “could account for,” “point toward,” and many others, is a statement of the limits of econometrics. The authors were wise to express themselves conservatively, because at this early stage of their examination it would be foolhardy to make more affirmative statements; yet their tentativeness doesn’t prevent them from adventuring deeper and deeper into econometric abstraction, laying out a mathematical ball of string that leads the reader over and through various logical hurdles and hoops to the desired conclusion.
For example, the ball starts here: “We begin by supposing that the economy evolves according to the following vector autoregression.”
They omit to tell the reader that the economy’s evolution in accordance with any particular vector autoregression is not yet established economic science. Remember: A chain is only as strong as its weakest link; and unfortunately, this particular econometric chain is made entirely of weak links.
Conclusion
The paper’s argumentation is an impressive piece of valid mathematics, but unfortunately the Fed’s conclusion is premature to say the least, and they might do better searching elsewhere for exculpatory proof.
The math may be sound; it is pointless to defend or challenge mathematical rigor when researchers are not even in agreement about the underlying suppositions and assumptions. Analyzing the math while ignoring the foundational weaknesses would be like test driving an elaborate 18-wheeler on an intricate and dangerous obstacle course without knowledge of the state of the tarmac.
Fisher and Quayyum might have written better science if they had whittled the report down to the four potential causes of our housing boom and concluded with the following sentence, with which 99% of all economists will agree:
“Determining the cause of macroeconomic fluctuations [through econometrics] is notoriously difficult because essentially all the variables of interest are exogenous — no single variable moves independently and drives movements in other variables.”
Amen.
Katy Harwood Delay, a columnist specializing in the fields of economics and government, lives in Los Angeles. Send her mail. See her archive. Comment on the blog.
