The puzzle of private sector research

Wed, 06/10/2009 - 9:41am

There's a lovely passage in John Le Carré's The Secret Pilgrim in which George Smiley explains why governments don't simply rely on open source information instead of spending gazillions on their own intelligence operations:  "governments, like anyone else, trust what they pay for, and are suspicious of what they don't." 

Oddly enough, in studying the global political economy, the sentiment often works in reverse in the academy.  Scholars, understandably, tend to prefer open source research while looking askance at private sector work that requires $$$ to unlock. 

I'm genuinely on the fence about this kind of question.  In writing about sovereign wealth funds, for example, I found the private sector stuff far superior on the empirics to the open source research.  The private sector stuff is also usually published before academics enter the breach (a good rule of thumb for aspiring IPE types -- if your literature review consists mostly of corporate research, then you are ahead of the academic curve on a new issue area).  On the other hand, the private sector work often lacked the analytical bite of scholarly work.  For some of it, I could not escape the sense that someone was trying to sell me something. 

I raise this conundrum because Martin Wolf's latest column is essentially a precis of a Goldman Sachs report that requires cashy money to read.  Wolf's summary: 

The paper points to four salient features of the world economy during this decade: a huge increase in global current account imbalances (with, in particular, the emergence of huge surpluses in emerging economies); a global decline in nominal and real yields on all forms of debt; an increase in global returns on physical capital; and an increase in the “equity risk premium” – the gap between the earnings yield on equities and the real yield on bonds. I would add to this list the strong downward pressure on the dollar prices of many manufactured goods.

The paper argues that the standard “global savings glut” hypothesis helps explain the first two facts. Indeed, it notes that a popular alternative – a too loose monetary policy – fails to explain persistently low long-term real rates. But, it adds, this fails to explain the third and fourth (or my fifth) features.

The paper argues that a massive increase in the effective global labour supply and the extreme risk aversion of the emerging world’s new creditors explains the third and fourth feature. As the paper notes, “the accumulation of net overseas assets has been entirely accounted for by public sector acquisitions ... and has been principally channelled into reserves”. Asian emerging economies – China, above all – have dominated such flows....

The authors conclude that the low bond yields caused by newly emerging savings gluts drove the crazy lending whose results we now see. With better regulation, the mess would have been smaller, as the International Monetary Fund rightly argues in its recent World Economic Outlook. But someone had to borrow this money. If it had not been households, who would have done so – governments, so running larger fiscal deficits, or corporations already flush with profits? This is as much a macroeconomic story as one of folly, greed and mis-regulation.

I'm pretty sympathetic to this argument, but I can't fully embrace it unless I can read the friggin' paper

Question to readers:  compared to academic work, how reliable is private sector research? 



Advertisement

 

"Cashy" ?

Does paying cashy money add to the paper's truthiness?

More seriously, I wonder whether the private papers will ever show up in anyone's archives. Surely the cashy value of the paper declines rapidly. Maybe Google, after they finish all the world's books, should start on the world's private research papers.

Believe it or not, that wasn't a typo...

but rather a Firefly -- well, actually, a Serenity -- reference.

I got it, and appreciated it,

I got it, and appreciated it, of course.

Transparency is key

Interesting comment, I definitely thought about this in college.

I am more likely to trust research conducted by members of academia than private sector work. Academic studies tend to be much more open and transparent about their methodologies.
How can we question Martin Wolf's column or the Goldman Sachs report when we are unable to determine how their research was conducted?

The methodologies employed by corporate research is usually unavailable to the public, so even if someone is ahead of the research curve, that does not necessarily mean the data should be trusted any more than existing data that can be scrutinized by the masses.

Emerging issues

I am not too confident either with corporate research, but they really highlight new issues. I have been struke by how they changed the perspective on emerging countries (BRIC anyone?).
By the way, what happened to your list of top ten books on IPE?
Never thought about doing one on emerging countries?

Commercial Open Source

Dan, a small point:

You write "Scholars, understandably, tend to prefer open source research while looking askance at private sector work that requires $$$ to unlock."

Commercially available data is still open source, though it all depends which sense of the term you mean: open source as in free of cost, or open source as a form of intelligence that's distinct from covertly acquired, protected, or processed information.

A basic assumption in any research is that transparency of method and rigor of quality control is the benchmark for reliable/replicable material. In that sense, I'd have to lean toward transparency of method rather than cash cost being a more important indicator on the reliability scale, between freely available information, information available for a price, and other forms of privileged data collection.

Ummm...it depends

"if you're literature review consists mostly of corporate research, then you are ahead of the academic curve on a new issue area"

I think you're absolutely right on this point. The nature of academic research and publishing means that scholars tend to be behind the curve. And low-readership, highly-specialized publications tend to pick up on trends really early on. Mark Hibbs reporting on the A.Q. Khan nuclear proliferation network for Nucleonics Week (which costs a gazillion dollars to subscribe to and caters mostly to industry types) is a notable example.

That said, because the intention of a lot of corporate literature is to sell products (I'm not considering trade publications here), data can be unreliable. There's also a tendency to jump the gun. This is particulalry true about security-related literature, including the stuff purveyed by the burgeoning private intelligence industry. Stratfor's reports, for example, are too topical or aggressive in their hypothesizing to make them appropriate substitutes for academic scholarship.

If you look at all research

If you look at all research works, not only the social science, but also the natural science and engineering areas, you are going to see that it is not a clear cut.

In some areas, corporate research has some edges since they are creating and experimenting the latest things. Even in finance, those "innovative financial products" were created and modeled by traders and investment bankers. Scholars only picked up those products later since there were going to be no public data if there were no actual transcations.

Dissident

There was no saving's glut. There was a penetration of the free-market system by illiberal economies, dirigiste economies, who are fundamentally non-market.

When China liberalises its capital accounts - we'll talk.

The Saving's glut theory - is one of those retrenched ideas, waiting for a Kuhnian shift.

I've spent some time tracking

I've spent some time tracking private research on Chinese marcoeconomics and one thing seems clear to me... As far as predictions go, they are just as bad as the rest of us in academia; but also worse, since they insist on making short-term (qoq and yoy) predictions.

A lot of energy also goes to try to rationalize past errors, instead of explaining things. As far as methodologies go, they are not very clear, especially about data series.

Another thing that detracts from the value of these publications is that the "politics" side in political economy issues is pretty much downplayed, or worse, reduced to gross simplifications.

For raw statistical data on things that are less subject to bias the reports are pretty useful though.