Endogeneity, Exogenous, Et Cetera

Louis-Philippe Rochon and Sergio Rossi have a very interesting article Endogenous Money: The Evolutionary Versus Revolutionary Views in the Review Of Keynesian Economics. I think it was written many years back and was in an unpublished form and has been published now. It is a nice critique of views of some Post-Keynesians such as Victoria Chick and also others such as Basil Moore. For instance, the paper quotes Moore’s view from 2001:

[w]hen money was a commodity, such as gold, with an inelastic supply, the total quantity of money in existence could realistically be viewed as exogenous.

Click the image to visit the ROKE website.

roke_cover

There are also some nice articles in a recent issue of JPKE on neoliberalism and the financial crisis.

Some gossip: The JPKE was initially supposed to have been called Journal of Keynesian Economics but it didn’t make it because the acronym would have been JOKE.

Also, Jayati Ghosh has written an excellent blog article on Thatcherism – the ‘triumph of private gain over social good’ (borrowing words from her).

Matias Vernengo has a recent blog post on the persistence of poverty in the United States. Which reminds me of an interview clip of Anwar Shaikh titled “The Sin Of Our Era”:

Back to formal matters.

What does it mean when an economist says words such as “endogenous”, “exogenous”? Most of the times, economists – mainstream economists – themselves confuse these terms and hence you see a lot of usage of these words in Post-Keynesian economics.

I was reading an article on econometrics by Fischer Black (of the Black-Scholes fame) titled The Trouble with Econometric Models

An exogenous variable is supposed to be a causal variable, if the structure of a model has economic meaning. In fact, it is usually just a variable that is put on the right-hand side of equations in a model, but not on the left-hand side.

Similarly, an endogenous variable is supposed to be a caused variable. In fact, it is usually just a variable that shows up at least once on the left-hand side of an equation

which is fair but there exists another language.

There is however another usage – that is in the control sense.

In an outstanding paper Federal Reserve “Defensive” Behavior And The Reverse Causation Argument, Raymond E. Lombra and Raymond G. Torto point out the following in the footnote:

Apparently no generally accepted concept of an endogenous money stock (or monetary base) has been defined. In statistical theory a variable is endogenous if it is jointly determined with other variables in the system. However, many monetary theorists have chosen to call a variable endogenous only if its magnitude is not under the control of policymakers. Such semantic problems have undoubtedly prolonged this debate.

For the money stock measure such as M1, M2 etc., there shouldn’t be any confusion. The trouble arises for things such as interest rates. For example, some economists may say that if inflation rises, the central bank may/will raise the short-term interest rate and it is endogenous while others will say it is up to the central bank to decide how much to change the interest rate, if at all. Such things lead to a lot of debate.

I like the latter usage (the control sense) but I think it is difficult to exclusively have the same usage.

The word “control” is also misunderstood. Here is a fine article on Wynne Godley in The Times from 16 June 1978 where he details on how misunderstood the word is:

Leading Economist Insists That You Cannot Control M3

(click to expand)

Leave a Reply

Comments are welcome, but not published—see comments policy. Required fields are marked *