Zero Hedge
All posts from Zero Hedge
Zero Hedge in Zero Hedge,

Mother Yellen's Little Helper - The Rate-Hike Placebo Effect

Via ConvergEx's Nicholas Colas,

Americans are increasingly likely to respond positively to a placebo in a drug trial – more so than other nationalities. That’s the upshot of a recently published academic paper that looked at 84 clinical trials for pain medication done between 1990 and 2013. Over that time, Americans reported an almost 30% incremental reduction in pain symptoms when given a sugar pill or other placebo as compared to a 10% reduction for in non-U.S. studies.  Why the difference? The paper’s authors suggest that drug advertising – only allowed in New Zealand and America – may be giving trial populations more confidence that a drug – any drug – will work.  Also a factor: drug trials are better funded now, and therefore have more participants and go longer.  All that may well spark more confidence in trial participants, even those taking placebos. 


These findings, while bad for drug researchers, does shed some light on our favorite topic: behavioral finance.


Trust and confidence makes placebos work, and those attributes also play a role in the societal effectiveness of central banks. That’s what makes the Fed’s eventual move to higher rates so difficult; even if zero interest rates are more placebo than actual medicine, markets believe they work to support asset prices.

I keep a mental list of underappreciated scientific developments of the 20th century, and near the top is the placebo. While you can argue that the roots and herbs of ancient societies were the first faith-based medicines, modern placebo research dates to a relatively recent 1955. That was the year Harvard research Henry Beecher published “The Powerful Placebo” in the Journal of the American Medical Association.  It was essentially a huge “You’re doing it wrong” to the pharmaceutical industry and showed that drug tests needed to be performed against a placebo and dual blind (neither subject nor researcher knows whether they were taking/dispensing a real medicine or a sugar pill). 

The only problem is that placebos are really stiff competition.  As Beecher noted, many subjects in non-placebo trials reported feeling better even when the drug in question was later found to be totally ineffective.  And over the years, researchers have found ways to make placebos “Work” even better: yellow placebos work great as antidepressants, red pills are “Uppers”, branded pills work better than generics, and a placebo painkiller given 4 times a day performs better when you take only twice daily.   The upshot of all this is that in highly competitive areas like depression medication, drug companies have to spend billions to produce products that beat the placebo. 

Medical researchers know that the placebo effect has risen over the last 15 years, but one recent paper attributes this to a very specific reason: Americans Dr. Jeffrey Mogil and his team at McGill University in Montreal looked at 84 drug trials dedicated to ameliorating pain caused by nerve damage.  Here’s a summary of their findings and working theories:

Since the mid-1990s, Americans in drug trials to assess pain medication have reported a 30% increase in the efficacy of placebos.  In Non-U.S. studies, that number is closer to 10%. Taken as a whole, therefore, American drug trials explain most of the increase in the global placebo effect.


The researchers at McGill do not have an explanation for this disparity, but suspect the changing scale of U.S. drug trials. We know that placebos tend to work better when patients trust their doctors (Kelley et al 2009). American drug companies have been extending the time involved in drug trials over the last decade as well as increasing their population size, in part to compete against the potentially wide variance of the placebo effect on smaller groups.  As it turns out, large drug trials where researchers get to know the patients well after weeks of contact are exactly the type of environment where placebos can shine.


The other idea forwarded in the press coverage around this study is that the U.S. is one of only two countries (New Zealand is the other) where drug companies can advertise directly to the population at large.  Pharma companies spend just over $4 billion/year on U.S. advertising, mostly on television ad buys.  Some of the study’s authors posit that this has a halo effect on drug companies generally, since Americans are unique in viewing these advertisements.  If called to participate in a drug trial, they may increasingly assume that they are going to get something new, innovative, and (presumably) effective.  Even if it’s just a sugar pill.

We see the placebo effect as medicine’s version of behavioral economics, that wing of the dismal science that recognizes the fallibility of mankind rather than assuming an economic actor will always behave as the models say they should.  If human psychology played no role in drug testing or medicine, you wouldn’t need a placebo option. If humans behaved like the wealth/utility maximizing animals described in economic models, you wouldn’t need a behavioral part of the discipline to explain why they often don’t.  The truth is human emotion gets in the way of both, so we need placebos and psychologists.

The central lessons placebos have for medicine is remarkably simple.

First, the patient/doctor bond plays a large role in the success of a placebo.  We know that from earlier studies and from surveys of doctors (one released two years ago in the U.K) which show they routinely prescribe placebos to their patients.  The doctors mean no harm, and the most often given reason for the practice was to ward off requests for medication inappropriate to the patient’s condition.


Second, as the McGill study seems to show, is that some populations will react to placebos more readily than others. The preconditions seems to be background messaging (all those ads no TV) and large/lengthy trials that encourage confidence among the patients in the company performing the trial as well as the researchers on staff.

To draw the analogy to economics, try this model on for size:

  • Central bankers around the world are the researchers, looking for effective treatments for slow global growth and sluggish price inflation.
  • The general population are the people engaged in the trials to see which treatments are most useful.

The American Federal Reserve ran the most successful trial with its Quantitative Easing program.  Asset prices went up, unemployment went down, and we don’t really have to care what parts of those outcomes were caused by the actual medicine of zero interest rates/bond buying versus the placebo effect of trust in Federal Reserve.  Now, the Bank of Japan and the European Central Bank are running the same experiment and hoping for the same outcome.  Whether the actual cure shows up or just the placebo effect is the big question just now. 

The challenge for the Fed now is that no one is quite sure how much of the “Cure” was medicine and how much was the placebo effect.  The first Fed rate hike will begin to tell the story, whenever that is, especially for equity markets.  It will all come down to how much investors actually trust the Federal Reserve to reduce the dosage of low interest rates and at what speed.  That is where the comparison to medical science is frustratingly inadequate.  Medicine cures, but when it comes to economics there’s always another illness coming around the corner.