Wednesday, November 11, 2009

Estimates and projections

All the financial writers on your publication's staff have been laid off, so your editor wants you--with your zero experience in financial reporting--to prepare a story about the direction of key economic indicators over the next two years. The focus will be on corporate profits nationwide, and it is your task to come up with some reasonable estimate of how profits will fare.

And no, your article may not consist merely of dueling quotes, as New York Times columnist Paul Krugman describes them, from economists with opposing views. You have to generate your own estimate, and then, perhaps, vet it with the views of experts.

Chill. First thing to keep in mind is that estimates--or in this case projections--are notoriously shaky things. The most solidly grounded projections can be upended by unforseen events: an unexpected war (think Iraq); the vaporizing of a bubble (housing, the debacle); implosion of some of the country's leading financial institutions (Bear Sterns, Lehman Bros.); a devastating earthquake.

But however common such cataclysmic events are today, you cannot approach your story with the view that they will always take place. You have to assume that the nation's economy does contain some degree of stabilization, without which no estimates, let alone projections, can be attempted.

So here's how to attempt it: We assume that corporate profits have a certain ratio (relationship) to Gross Domestic Product. Therefore,

1) Get an official listing of corporate profits over the past few years.
2) Get an official listing of GDP over the past few years
3) Get an official projection of GDP over the next few years.
4) Divide existing profits by existing GDP over the past few years. This gives you recent annual percentages--your ratios, in other words.
5) Now note if these percentages follow a slight pattern. For example, do they decline each year? Increase each year? Go up and down?

All this is easier than it sounds. Let's start with steps one and two. Both recent corporate profits and recent GDP figures are available from the Commerce Department's Bureau of Economic Analysis. Click on this site, and go to Table 11. On the top line, you'll see that corporate profits for the past three years were:

2006: $1,608.3 billion (or more than one and a half trillion dollars).
2007: $1,541.7 billion.
2008: $1,360.4 billion.

In other words, recent profits in general (as of this posting) have declined over the past three years. No surprise there, given our deep recession. (Ignore the fact that some companies made out like bandits during the same time period.)

Now go to Table 9. You'll see that GDP was:

2006: $13,398 billion.
2007: $14,077.6 billion.
2008: $14,441.4 billion.

GDP rose slightly. (If only jobs and wages rose along with it! But stop digressing.)

Now divide each year's profits by that year's GDP. Use a calculator, or better still, a spreadsheet program like Excel. Here's what you'll find: In 2006, profits were twelve percent of GDP. In 2007, they were eleven percent. And in 2008, they were 9.4 percent.

In other words, recent profits appear to follow a slight downward pattern in relation to GDP, dropping by one to two percent of GDP each year.

You're almost done. Now you need to apply this pattern to the GDP projections, which you can get from the Congressional Budget Office. The top line of this report shows that the CBO projects GDP for 2009 at $14,163 billion; and for 2010 at $14,570 billion.

So take, say, 8.5 percent of the 2009 amount. Why? Well, it follows the pattern: it's nearly a point below the 2008 figure. (You can tinker with this percentage a bit, after you talk to an economist or two.) So you get $1,203.85 billion projected profits for 2009.

Now take, say, 7.2 percent of the 2010 GDP figure: you get $1,049.04 billion projected profits for 2010.

You now have your own projections. Are they accurate? Well, many factors affect profits, of course, and anything can happen to disrupt the pattern. But other things being equal, these projections are, indeed, reasonable assumptions.

No comments: