Show me the money

A James Alexander post

I am not a wonk, and barely an expert. This is a bit wonkish, but still worth reading for the non-expert. Mark Sadowski inspired it!

In Tony Yates’ reply to my earlier post he cited two academic studies that would presumably show me that what financial markets know to be true, was in fact false.

Yates and the “long and variable lags” brigades who flourish inside our central banks and elite universities remind me of quack doctors in the world before the discovery of the circulation of blood. They prescribe remedies for which they have no agreed scientific basis, but don’t really get tested because if the the patient recovers they win, if the patient dies it was because the disease was proven uncurable. We only find out the effects of their hit and miss cures at the end of the treatment.

For anyone wanting a quick read on just what a mess modern, consensus, macroeconomics is in the Wikipedia entry on the standard paradigm. “New Keynesian” reads like a discussion between two competing schools of seventeenth century apothecaries. No-one has a clue, but they definitely know the other side is wrong.

Modern doctors prescribe cures to which they generally know the results. Start this course of drugs now and in a certain period of time you will be cured. Setting the course is everything. If doctors didn’t know ibuprofen stopped most headaches they wouldn’t prescribe it. The long and variable lag from the taking the pill to the headache going is completely irrelevant.

Curiously, the first paper Yates cited (Christiano, Eichenbaum and Evans) is actually in agreement with the supposedly “utopian” Market Monetarists. It’s hard to know why Yates brings it up, but thanks anyway. The paper contains a crucial quote:

“To actually implement a particular monetary policy rule, the growth rate of money must (if only implicitly) respond to current and past exogenous shocks in an appropriate way. This is true even when the systematic component of policy is thought of as a relationship between endogenous variables, like the interest rate, output and inflation.”

Put simply, monetary policy changes have to act quickly on money, on the money side of the economy itself (the MV bit of MV=PY). If it isn’t acting, or expected to act, it isn’t working.

Market Monetarists say you can tell if it is working instantaneously by looking to see if the size of the money economy is expected to grow. Scott Sumner’s great contribution is to suggest a “money economy” growth futures market, or NGDP futures market. It would be the equivalent of all the best medical brains putting their money where their mouth was, or reputations on the line. Doctors in medicine do this today, everyday, by earning money through proven, successful treatments. Setting courses, more or less knowing the results – not much waiting around to watch the long and variable lags. They act with confidence in the outcomes of the courses they set.

In the absence of futures markets we are forced have to use somewhat unsatisfactory substitutes like market forecasts of inflation (in the US TIPS spreads), or other forward-looking instruments like stock market indices and bond yields, or FX rates that incorporate the expectations of one currency’s inflation prospects vs another (amongst other things).

The market’s reaction is the policy. When the market reacts to unexpected data points or new news in FOMC speeches, decisions, leaks, minutes, etc, then that tells us how the policy is changing. Immediately. Instantaneously. With no lag. Unexpected silences from the Fed in the face of big market moves can also be telling, as now. It is harder to interpret than unexpected news, but it is still there.

Yates directed me to his own co-authored paper but it seems as weird as most modern macro in that it has no role for money. The equivalent of those quack doctors working without knowledge of the circulation of the blood. When writing about monetary policy with such authority as Yates and so many macroeconomists like him, the plain man would expect money to figure in their highly technical models of the economy, but it doesn’t. No wonder Yates and so many central banks like the Fed, the BoE, the ECB who use the same moneyless-model have proven so hopelessly wrong during the crisis. Their model is literally and metaphorically bankrupt due to the absence of the key measure of monetary policy, money or the money economy itself.


7 thoughts on “Show me the money

  1. To me, the interesting thing about this topic is that if you ask an economist–any economist–for a starting point for economics, they rarely compare. In other words, unlike a science or even social science, there is no consensus story or narrative behind the field. This is profoundly disturbing. If you get a 3 day course on economics foundations and essential truths, no one has either a macro or micro story except the behavioral economists. Very odd.

  2. James, I’ve used a similar medical analogy myself on more than one occasion. For example, here.

    But in my case, as an outsider to both economic and medical science, I look around at the profound disagreements in the field and say “The whole field of macro looks like medical science used to in the 13th century… ALL of it!” It looks like arguments between blood letters, cuppers and ear candlers to me. It’ll be a “miracle” if any of you turn out to be right about anything.

    As I commented to Nick, as far as I know, Market Monetarists and NGDPLT will be the best thing since the wheel (in fact I’m a fan of the MM story: it sure *sounds* good), but it still seems to me like macro is a few centuries early in offering ANY advice with any kind of conviction behind it. Why not show that your theoretical framework and the models derived from it can pass 1000s upon 1000s of falsifiable empirical tests/predictions (tests designed to distinguish MM from competing theories) first without any failures (like science demands in other fields) and without having to add any “epicycles” to the theory before moving confidently to policy advice? If macro were at all mature as an observational science, then I can’t imagine there’s be so many fundamental disagreements in the field (as rlanham1963 points out above).

    Let me ask you this (as I’ve asked so many others): What concrete examples of conceivable empirical evidence would convince you that MM is wrong? If you can’t specify very precisely the conditions under which it would be clearly falsified, then is it really science?

    Regarding money: again for all I know, you might be absolutely correct (again, your story sound intuitive and satisfying to me). However, it’s not always clear (to me) that just by “looking at it” what is most important to include in models is apparent. For example, the ideal gas law depends on almost zero characteristics of the gas molecules in question: the only thing that really carries over from the “micro” properties of individual gas molecules to the macro properties of the ideal gas law is the number of degrees of freedom the molecules can move in. Other than that, we just rely on the fact that there are LOTS of gas molecules experiencing movement in all those degrees of freedom for the gas law (and statistical mechanics in general) to work. it doesn’t depend at all on the structure of the protons, neutrons or electrons making up the atoms in the molecules. Likewise, when we zoom in and look at an individual molecule, it no longer has the macro properties of the gas (properties like pressure and temperature). How can we have confidence that this all hangs together? The theory makes predictions that have been checked over and over without ANY failures. What testable predictions follow from your statements about money? How will we know if those statements are wrong?

    • Tom.
      Good to see you are sympathetic to MM.

      Actually, I get your challenge to economics most evenings from my physicist son over the dinner table. There never will be physics like tests in economics, repeatable experiments and the like. However, I switched to MM when the quadrupling of the monetary base didn’t cause any inflation, my earlier prejudices were falsified. I would switch away if NGDP Targeting were tried and led to a recession or a runaway boom.

      When Mark Carney merely raised NGDP Targeting as a possibility in December 2012, six months before officially taking up the BoE job, it electrified markets and boosted the UK recovery. I recommended to an official close to Draghi that the ECB President should do the same, just as an experiment. Sadly, the message hasn’t got though yet, but it would be interesting. I predict it would boost markets/weaken the Euro/raise NGDP growth expectations.

      MMs have been alone by consistently saying it is too early to raise rates in the US, we have been tested this last week as the FOMC signalled it was even nearer the time to do it. What was the result? Taper Tantrum 2.

  3. Lorenzo
    Have read your piece from 2013. Very good. I will link to it in future.

    I get your point about somewhat less than half of all transactions involve money, but I wonder whether the personal transactions really are transactions rather than just life. “If you clean your teeth I’ll read you a bedtime story” isn’t really a transaction. If you discount life but include FX transactions then more than half of all transactions will involve money.

    The most remarkable thing I found in When Money Dies was how the Germans were at full employment during the hyperinflation and actually thought the economy was doing rather well. Even if money savings were destroyed those owning equity, mostly private holdings, were doing well.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.