Modern Economic Theory Is Rotten To the Core

Story Stream
recent articles

There is often no need to be reminded how small the margins for error can narrow, as in complex systems the explosion of error terms thought to be just random or unexplainable noise can be beyond catastrophic. That is the essence of chaos theory, or fractal geometry, but such observations are not limited to clinical study of mathematics or complex systems. Error is as much a part of real life as it is useful in statistics and critiques of them. Innovation, after all, is the product of error and the occurrence of "tail events", but there is much about negative errors that can be misunderstood, particularly when it occurs as a result of intentional intrusion.

Take for example the drug Rezulin. Most people, I believe, will have no memory or recollection of the enduring damage left by the radical diabetes medication. The string of events that led there actually began by a simple number changing from 78 to 48. While the mathematical reconstruction here was not random, it was, in fact, highly improper and seemingly intentional (I haven't followed the numerous court cases all the way to conclusion so I cannot speak about any factual findings of intent), that small numerical alteration transcended its narrow use and original applicability.

In the middle of the 1990's, pharma giant Warner-Lambert was pushing for Rezulin to gain fast-track approval from the FDA. They claimed, through studies, that initial indications were of great benefit to diabetes patients without as much attendant liver damage as had occurred in prior medications treating the disease. As anyone with diabetes will tell you, and I am speaking for myself, liver and kidney function is always among the first considerations for not just evaluating treatment options but looking toward longer-term health. Thus, Rezulin offered a potential evolutionary advance into what was then just growing into a significant problem.

By the time the drug was pulled from the market in March 2000, it had gained almost $2 billion in sales. But there were glitches almost from the moment Warner-Lambert had received fast-track approval, as the first acknowledged death, due to liver failure, was found as far back as October 1997. All told, by the time of the drug's removal there would be some 90 accepted liver failures due to Rezulin usage, including 63 deaths and a further 7 organ transplants. The monstrous wake of the shocking failure left billions in lawsuits that have cost more than the revenue brought in from the effort.

As I said at the outset, the reduction from 78 to 48 allowed all of that to proceed and fester. Even before the drug was released to general medicinal prescription, studies had indicated possible liver conditions. When the drug was first approved in March 1997, there had already been 78 cases of patients reporting some stage of liver damage. When public information about the first recorded case of liver disease was about to be released that fall, Warner-Lambert proactively took what was then the lauded step of attaching a warning label telling doctors to monitor for, and remain vigilant about, liver problems.

According to a Los Angeles Times account, on October 23, 1997, Warner-Lambert execs, including the head of the diabetes unit, met with FDA officials to discuss the liver label. The 78 cases found in clinical trials was purportedly never brought up at that meeting, only in subsequent written communications were 48 cases revealed to the FDA.

The reduction from 78 to 48 has, again to my knowledge, never fully been explained. No rationale was provided other than some notes and referrals to hand-written "rules." The practical effect of excluding those 30, however, was to reduce the statistical count of potential liver problems from an alarming 3.1% (out of 2,510 participants in the drug trials) to just 1.9%. At 1.9% the warning label was deemed appropriate, whereas at 3.1% more severe action would likely have taken place (maybe even including a public FDA warning). Even still, the drug was taken off the market in the UK in 1998.

Of course, over the next three years as sales turned Rezulin toward blockbuster status, more liver damage became known and a backlash ensued, from both outside and within the medical community. Being the times as they were, Warner-Lambert sought reinforcement in various forms, including advertisements and endorsements. One of those "endorsements" was a paid position as sponsor for the "Diabetes Center" of the now-infamous At the time, Dr. C. Everett Koop was known as "America's Doctor" owing from his high profile and very much outspoken and independent position during his time as Surgeon General of the United States.

What looked like a surefire partnership rapidly devolved into chaos for both. The online portal for "health content" had been launched to great notoriety and acclaim, during what is now known as Web 1.0. It was a wild period with all sorts of open sky claims and endless straight-line extrapolations upward and onward, forever rising. had gone public in June 1999, having raised an impressive $88.5 million. In no time, the site attached to the company had become the #1 health care portal with as many as 1.5 million unique visitors per month. That was a big deal in 1999, as paltry as it sounds from a Web 2.0 perspective, and it represented a kind of quasi-currency that was used to "attract" so much "capital."

However, to gain that traffic status actually cost real currency. Only a month after the IPO launch, they "partnered" with AOL to be the "premier" health content provider across what was then five of the "leading" internet brands (that now are just bits of arcane web history): AOL, CompuServe,, Netscape Netcenter and Digital City. The terms of the agreement to drive all that traffic were enormous, with paying AOL $89 million over four years, plus giving AOL both fully vested and performance-based warrants to purchase stock in the future. It seemed like a good bet given that's stock price had risen from its $9 offering to over $45 at its peak near the AOL deal.

The company never made it far, as only eight months later's auditors, PricewaterhouseCoopers, filed a note (during the same month that Rezulin was pulled) with the company's quarterly 10K that expressed sincere doubts about the company's ability to remain a going concern. The dual slash to credibility was just too much for the strained business model to bear. Management acknowledged the substance of the warning, without saying anything about Warner-Lambert at all, but pledged vociferously to obtain new rounds of financing through additional equity subscriptions or even venture funds. It never occurred, and operations scaled back significantly and disastrously - without "traffic" there was no "value", and so no ability to gain further finance. The end came in December 2001 as the website shut down for good.

You have to wonder, sifting through all that wreckage more than a decade afterward, what might have happened had that 78 remained a 78. Rezulin probably drops off in sales and with it the company's usage of resources to grow them, and probably doesn't attract AOL as without the need to sponsor the website company looks far different. The downstream implications beyond just those immediate counterfactuals "prove" far more difficult to discern relatively plainly, but we can reasonably surmise that so many resources might have found much better use elsewhere. Unfortunately, that includes lives that probably would not have been lost, and health issues that complicated living for more than a few others.

In strictly economic terms, though, there is a clear and even measurable downside to having undertaken the effort. Even though it would have been remarkably costly to go back to the drawing board after Rezulin's problems became so apparent, it proved far more costly to not have done so. In the case of, it was a failure all too common for that era - the company filed for $9.4 million in revenue against $56.1 million in losses in its 1999 reporting that raised its auditor's deathly ire. was no outlier, and was really just one of the more prominent failures among too many to recount. The commonality was the imposition of web "traffic" as some kind of substitute for actual revenue and fiscal results. Normally, that would not be so much as a consideration for anything other than pleasantries, but at the time that was taken to "market" as if it were real and valuable. Traffic, and not even traffic but simple promises of it at some undetermined point in the distant future, was enough to generate billions in actual cash flow in the form of equity offerings. Some of that was certainly unrealistic expectations, some misunderstanding of the brand new era of online, but that does not encompass the full "froth" of the mania. Far more was destroyed than created, but what we are learning now, and have been for twenty years, is that the intangible destruction may be far, far more costly than what was seen and described.

In the case of economic policy, however, current orthodoxy dictates that no amount of economic activity is "bad." That applies equally to the fiscal side as the monetary side, as the Keynesians and the monetarists are united on that front (as they are in almost every important fashion, including the utter adherence to government means as the only method for economic growth and advance). In the words of Paul Krugman,

"This is the kind of environment in which Keynes's hypothetical policy of burying currency in coalmines and letting the private sector dig it up - or my version, which involves faking a threat from nonexistent space aliens - becomes a good thing; spending is good, and while productive spending is best, unproductive spending is still better than nothing."

That sentiment is taken as gospel because of the idea of "pump priming" combined with monetary neutrality. In essence, spending for the sake of spending, or activity for the sake of activity, is better than "allowing" resources to remain idle. The implication, what Krugman and his strain infer, is that there is no cost to creating artificial incentives to simply "do something."

In a more extreme example, you can imagine a conforming case whereby such "stimulus" is enacted through something like loans for pizza parlors. So many people are convinced that they can cook and prepare pizza far better than what is offered, so why not give such potential "capitalists" so much "free money" in the form of zero- and low-interest loans at extremely generous terms to find out? That would create a rush of "investment" in machinery and retail space as to foster the theory of spending begets spending, thus providing the economy with a direct, short-term boost.

There would be an increase in the amount of ingredients for dough and sauce, as well as toppings and any sides and drinks, that we can reasonably assume those suppliers would likely generate further incremental activity all the way down the supply chain.

And so it would be, a rush of activity that would look like an advance in the economy creating broad-based downstream impacts (not taking prices into consideration). Intuitively, however, we know exactly how it would end and it would not take long to achieve it. Taking no account of anything like profitability and sustainability, the saturation of pizza joints would not only cannibalize each other, they would do so to existing establishments of all types. After the initial rush of activity, these very projects would be beset and plagued by widespread losses failures in very short order.

The orthodoxy says that does not matter; all that matters is activity for the sake of activity, a generic conception of the economy that councils nothing by way of wealth and true capital. While my parable seems to be a case of reductio ad absurdum, we have actually seen its like in action - having described an example of exactly that in the paragraphs above (and then again in mortgage and housing). was not a unique outlier of wasted resources, but rather far too typical -,, (Disney's massive waste), GeoCities (Yahoo! bought it for $3.6 billion in 1999!), and everyone's nostalgic favorite and its Super Bowl star sock puppet. Billions upon billions were wasted chasing "traffic" without any exact means to extract revenue, let alone profits. The immediate results of all that, however, were as righteous in the orthodox conception as anything else.

The collective failure of the dot-com bust produced a recession that was perhaps the mildest on record. In fact, viewing the consumer side alone leaves little impression of a recession at all, yet there was the quite severe downturn in capital spending (that in some places still has yet to be fully surpassed even almost fifteen years later) and employment that dragged out into the first major instance of a "jobless recovery."

Even the staunchest allies of IS-LM economics, the later Keynesians that adopted econometrics, now acknowledge that discrepancy between the consumer and business sides in the early 2000's as rolling asset bubbles. Krugman again:

"We now know that the economic expansion of 2003-2007 was driven by a bubble. You can say the same about the latter part of the 90s expansion; and you can in fact say the same about the later years of the Reagan expansion, which was driven at that point by runaway thrift institutions and a large bubble in commercial real estate."

So if I read this historical context correctly, then spending and activity for the sake of spending activity leads to no longer-term economic problems but somehow the economy needed another bubble shortly thereafter to achieve even unsatisfactory levels of growth? I think there is a huge discrepancy and contradiction here that is quite obvious outside of ideological rigidity, as the waste of resources in the first place might have something to do with the acknowledgement of what followed. Some might quibble with my use of the term "needed" as a qualifier in the first sentence of this paragraph, but that is the contention now offered by these same economists in both fiscal and monetary camps.

The current state of economic theory posits a secular stagnation befalling not just the US economy, but globally. The predicate position of secular stagnation is a negative natural rate of interest, which has purportedly been falling for at least thirty years. Under the academically simplistic IS-LM framework, the economy "can't" grow unless the visible rate of interest in the economy is actively reduced below that natural rate. So if the natural rate is negative, what follows is that the actual rate must be more deeply negative to achieve any sort of liftoff - thus QE's roll forward with regularity and an eye toward ramping up "inflation expectations."

However, the downside of so much negative interest rates, natural and real, is that they are the monstrosities at the heart of rolling asset bubbles that Paul Krugman so helpfully notes. Even Janet Yellen's Fed has seemingly adopted just such a stance, though they will never admit as much. Thus, the current incarnations of QE, the third and fourth in the string, are headed toward pre-programmed exits because they are observed to be diminished in their capacity toward generating sufficient "inflation expectations", which does little toward rectifying the discrepancy (what is assumed, anyway, by the orthodox practitioners) between observed rates (real or nominal) and that natural rate assumption. With so little benefit in the IS-LM condition, the costs of such interference in risk pricing become far more paramount (that is the reason for Yellen's penchant for mentioning biotech stocks, and other FOMC participants talking about "reach for yield" as well as the insane party taking place in junk bonds and leveraged loans).

To explain exactly why the natural rate has stooped so low so long has taken on a sadly comical turn. Some hypothesize about demographics, though demographics have never been a serious problem despite ups and downs in our history before. Economist Tyler Cowen even offered, seriously, that there has not been enough war in the latter part of the 20th century and into the 21st. I believe he also offered the "explanation" that innovation has become much more difficult now that all the "low hanging fruit" has been plucked (without much by way of suggesting why there is no more low hanging fruit).

The depths of convolution that economists wish to take to try to square their circle, to match the ends of the argument over asset bubbles and secular stagnation, is itself a wonder to behold. The tortured logic is required because the position is rotten from its base - that spending for the sake of spending has no lingering downside, that "idle" resources are better wasted than left idle.

The role of monetary policy under such wasteful conditions is as a steepener of the failure curve - "reach for yield", whether in the form of junk bonds and leveraged loans or dubious IPO's of websites with nothing but a general idea for web traffic, keeps waste wasting far longer than it "should." In the course of doing so, the spread of failures in the inevitable shakeouts as these unprofitable enterprises gain final reckoning is compressed significantly into huge crests. That is what we see as bubbles, as their end is nothing more than that final reckoning of prior waste as the spell of monetary influence inevitably blows backward all at once. That was 2000-01 and 2007-08.

It has been those waves of failures, rather than a steady trickle of them, that has seemingly depressed the natural rate of interest. It is difficult to expect much long-term deployment of true capital when your base case for expectations is exactly that, bubbles generating wasteful behavior that eventually coalesces and clumps into a ball of nastiness that extends far beyond the initial reach of generically altered activity; what some call throwing the baby out with the bathwater. Under such uncertainty, the allure of "easy money" in the financial sphere is too powerful, and thus actually suffocates any nascent impulse to invest productively (that counts for not just corporate offices, but also the speculator culture of unbridled lust for "easy money" that has entrenched itself).

What we can learn from and Warner-Lambert's foray into suppression of 30 cases is that though the numbers and margins might be small there is a massive difference between activity and wealth. None of what either firm did was ultimately profitable, which is what a capitalist economy needs to grow over more than short bursts. Since wealth creation is axiomatically tied to profitability, wasting resources in broad clumps indisputably ruins profitability, meaning wealth. This is all about proportionality, as even a healthy economy will create waste and failure, but we need not seek the means and ends by which to augment that failure and turn them into tsunamis. The economy needs nothing to do with bubbles, only to be allowed to once more create wealth free from that destructive interference.


Jeffrey Snider is the Chief Investment Strategist of Alhambra Investment Partners, a registered investment advisor. 

Show commentsHide Comments

Related Articles