Pages

Thursday, March 12, 2015

Note To Reporters: You Cannot Analyse A Sovereign Like A Corporation

Reporters and (apparently) readers have a hard time differentiating between the different fields within finance, and so they happily treat comments made by people like internet analysts (!) as representing "what the markets think" about government fiscal ratios. Since bad fiscal analysis is more exciting than good analysis, that is what appears in the media.

Please note that the difference between corporate financial analysis and the analysis of governments is found in the book Understanding Government Finance (link).

Very Quick Justification


I am not going to give a long explanation of why you cannot analyse a "sovereign" (see comment below) like a corporation here. I will instead offer a simple example of how the logic in one area is diametrically opposed.

Let us assume that a country is expected to have strong, steady growth (let's say 7% nominal GDP growth) over the coming 5 years.

  • A corporate financial analyst would expect a (successful) domestic corporation to ramp up spending, as its revenue should rise in line with nominal GDP.
  • A (good) sovereign analyst would say that this is a good time for the sovereign to cut spending, on Functional Finance grounds (primer).

In other words, the standard logic used by corporate analysts is exactly backwards when applied to government finance.

Another major difference is that the central bank can ensure the liquidity of government debt as it is rolled over, whereas (non-bank) corporations cannot. During a financial crisis, a sovereign has no problem with rolling its debt, whereas nonfinancials get their credit lines cut. (At this point, Modern Monetary Theory comments will appear about a "currency sovereign" not facing default risk; this elimination of "rollover risk" is why this is true.)

To be clear, countries that borrow in foreign currencies (including the inmates of the euro area) and sub-sovereigns (such as Canadian Provinces) start to look like corporations in terms of default risk. But I worked for a period of time as a Canadian Provincial Bond analyst, and even there, one would get limited insights from the techniques used by corporate bond analysts,

Examples


Japan is the mother lode of examples of this. Two recent examples include:

But it is not solely Japan. A classic example of analysing the government like a corporation was the analysis given by Mary Meeker, an internet equity analyst during the bubble of the 1990s. In  Mary Meeker's Definitive Guide To The American Public Debt Crisis, we find:
Kleiner Perkins partner Mary Meeker analyzes America as a corporation in an epic presentation: USA Inc.: A Basic Summary of America's Financial Statements (pdf).
Meeker says USA Inc. is on a parallel course to GM before its bankruptcy.
It seems to me that prognostication did not age very well.

Bad Fiscal Analysis Drives Out Good


Although it may look like I am cherry-picking some hapless cases, I would argue that there is an underlying principle at work. I have never seen a media article that could be summarised as follows:

[Corporate Analyst/Portfolio Manager] X has looked at the fiscal ratios of [currency sovereign Y], [uses a metric that is applicable in corporate finance] and concluded Z, where Z is a statement that is not completely silly.

To be clear, I am very sure that there are corporate analysts who have entirely sensible things to say about government debt ratios. But since sensible statements about government finances are actually fairly boring, they are unlikely to be aired in the media. Editors want exciting headlines, and you can only get there by using some fairly dubious analysis techniques. For example, there are a lot of good Japanese rates strategists who publish articles for the investment banks; but I would be hard pressed to remember any time that their views were broadcast by the public media.

Additionally, I am not counting statements about bond yields: since they go up and down all the time, in principle an analyst has a 50% chance of being right.

If a reader objects to my generalisation, feel free to supply a counter-example so that I can apply my skeptical eye to it.

See Also:

(c) Brian Romanchuk 2015

2 comments:

  1. Yes, it would be nice if economics commentators understood the difference between macro and micro. Coincidentally I'm writing a post for my own blog about similar mistakes made by the UK's "Institute for Fiscal Studies" which I'll publish in the next day or two.

    ReplyDelete
  2. "...since they go up and down all the time, in principle an analyst has a 50% chance of being right." LOL, you got that right! Just look at what's happening in oil right now. I turn on CNBC and every week someone is trying to call the bottom. Eventually one person will be right and they can write a book and tell everyone how great they are at picking tops and bottom.

    ReplyDelete

Note: Posts are manually moderated, with a varying delay. Some disappear.

The comment section here is largely dead. My Substack or Twitter are better places to have a conversation.

Given that this is largely a backup way to reach me, I am going to reject posts that annoy me. Please post lengthy essays elsewhere.