Saturday, 12 February 2011

The Tinbergen Rule.

The Tinbergen rule was formulated by the economics Nobel laureate, Jan Tinbergen. There seems to be some confusion as to exactly what the rule consists of, but it is often stated as something like “for each policy objective, at least one policy instrument is needed”.

For example to deal with the problem that the elderly need support (a policy objective), some sort of pension policy is needed (a policy instrument).

The Tinbergen rule is stated by Dr. Derick Boyd* very much in the above form, i.e. “there should be at least the same number of instruments as there are targets”.

I suggest Tinbergen made a mistake here (if the above is a fair summary of his rule). Reasons are thus.

It cannot make sense to have more than one instrument address a particular objective because one of the instruments has to be better than the other.

Of course it is always possible that one instrument deals with a particular ASPECT of an objective better than another instrument. But in this case, doesn’t that aspect become a separate objective?

For example, if one pension scheme covers funeral costs and a second one doesn’t, then the former is arguably superior. But “funeral costs” is a separate objective to “costs of supporting the elderly”.

In other words, the principle should state, “for each policy objective, one policy instrument is needed, and one only.”

Thoughts, anyone?

A few minutes of Googling will produce plenty of material on the Tinbergen principle, but here are a couple of “starters”

* The Theory of Macroeconomic Policy, East London Business School, (available on the internet).



  1. The Tinbergen Rule was formulated specifically in the context of macroeconomic policies in an open economy. If you have two policy goals - stable prices and low unemployment - you need two independent (and effective) policy tools: say monetray policy for stable prices and fiscal policy for low unemployment. But in an open economy there is another policy goal - external balance. That may take the form of maintaining a fixed exchnage rate, btu if a nation does that it must devote monetray policy to achieving that goal, because exchange rate interventions cause the money supply to change (the country can try to "sterilize" such interventions, but that won't work in the long run). That leaves fiscal policy to achieve internal balance. Not a problem if the economy is experiencing falling pices and high unemployment, but a big problem if it is experiencing stagflation, in which case it must find another policy tool (wage and price controls?)

    but if the country uses a floating exchnage rate, then changes in the exchnage rate achieve external balance automatically (assuming the Marshall-Lerner conditions are met), and the nation still has both monetary and fiscal policy to devote to internal balance. Problem: fiscal policy tends to be less effective under flexible exchnage rates, and the problem is worse the more easily capital flows between nations>

  2. Hi there Anon, Whatever mix of policy instruments is chosen, there is always the problem that adjusting one instrument will adversely affects the objectives of another instrument. But that basic truism is not a weakness in the Tinbergen principle: the latter just says one should just choose the BEST instrument for each job and live the undesirable side effects.

    The reason things work out badly in your example, if I might suggest, is that you’ve chosen poor instruments for each objective. E.g. you choose “monetary policy for stable prices and fiscal policy for low unemployment”. I think that is a defective choice because both monetary and fiscal affect demand, inflation and employment. Monetary and fiscal are too similar to be allotted different jobs. In fact in other posts on this blog I have suggested that the distinction between the two should be abolished. (Basic reason being that employing either on its own is distortionary.) This suggestion that monetary and fiscal be merged is also advocated here:

    Also, you choose fixed exchange rates to achieve external balance. That is a very poor choice, seems to me. I fact fixed exchange rates will exacerbate external balance. In contrast, FLEXIBLE exchange rates ought to achieve external balance (if exports and imports respond enough to devaluations or revaluations).

  3. Ralph: Ah, good. You've discovered Tinbergen. If monetary and fiscal policy both affect AD, but if they have different effects on other variables of interest to the policymaker, you don't ignore one of those two lever.

  4. Funny thing is, I was just Googling "Tinbergen", precisely because of the recent debates on monetary vs fiscal, including our debate, and Google brought me here!

    The "novel" argument I was advancing, which you posted about a few weeks back, was really just a version of Tinbergen.

  5. Nick, Yes, if it can be shown that monetary policy is much better than fiscal (or vice versa) at dealing with specific “policy objectives” then there is an excuse for separating monetary and fiscal. But I’m sceptical.

    For example, an interest rate rise combined with a fiscal boost might ameliorating a house price bubble, but 95% of the experts failed to see the seriousness of the pre-crunch housing bubble, so that argument rather falls flat.

    Monetary might work faster than fiscal (or vice versa), but I don’t know if any brilliant evidence to support that idea.


Post a comment.