What?

It's a simple question ,What.  But it is not always easy to know "what" something is.  And I don't mean just non-descript lunch meat.  For example, some of the basics of modern economics depend upon measures, indicators, and measurements that are not always "correct" in terms of measuring what we wish to measure.  An example is Ben's discussion of Stiglitz's book on rethinking whether GDP is the best indicator of performance and development.  What is the actual production of an economy?  This definitional question can be surprisingly difficult to answer.  (I would thus note that definitional questions are sometimes good ones, notwithstanding my "mainstream social scientist's" view that the most appropriate questions for our purposes are causal in nature.)

On a previous occasion, I noted being struck by the folks at the IMF who said they felt like the institution had got things precisely backward in the Asian crisis.  That was an error - understood in hindsight - based on an understanding of economic dynamics.  But even more fundamental is if economists don't know what's actually going on in an economy.  What is the actual amount of activity?  What is the amount of output?  What is the amount of production?  What are the prices?  What is being imported, and what is produced domestically?  What?  What?  What? 

These may seem relatively simple to answer, but examples abound that show we don't even know what the current indicators are.  A common example is the need to revise various economic estimates - such as GDP or job gains and losses - a couple of months after initial estimates are sent out.  At some level, this is not to be too much lamented, perhaps: we know we rely on estimates, and accept their veracity at our peril.

But sometimes a presumed measurement becomes conventional wisdom, and then affects policy assumptions.  Which can be dangerous if the underlying measures are wrong. Take this article from the Washington Post.  It says that US manufacturing has done much worse in the last decade than was feared.  Conventional wisdom of late (no doubt fueled in part by economists for whichever party holds the White House, as the article notes) had become that manufacturing is actually on the upswing in the U.S., on the back of much greater labor productivity.  But this article holds that such claims were based on wrong estimates of what is actually being produced in the U.S.  In short: our top economists have been giving us the wrong numbers for what we actually produce in this country.  D'oh!

Take this new finding with a grain of salt as well, and note that it does not eradicate all of the growth in labor productivity.  On the other hand, if you have become convinced that the U.S. will do fine in manufacturing going forward, and are inclined to suggest policies accordingly, you will want to make sure you know "What" you are talking about.  The implication of this article being, in part: Not that most economists actually do.