Growing the GDP with Promises


Back in April, this line from a Financial Times article on Bureau of Economic Analysis (BEA) revisions to the method of calculating gross domestic product (GDP) caught my eye:

Deficits in defined benefit pension schemes will also be included [in GDP] because what companies have promised to pay out will be measured, rather than the cash they pay into plans.

Page 21 of this PDF offers an almost-plain-English explanation of the change.  If I’m understanding correctly, the BEA will essentially be adopting the accrual and “normal cost” method that governments (and other defined-benefit employers) use to calculate their contributions and liabilities.  In practice, the present value of the promised benefits will count as employee compensation, regardless of whether the money’s there or the investment return rates are plausible, and then be accounted as a personal investment.  Then:

Because BEA measures government consumption expenditures using input costs, the revisions to compensation of both federal and state and local government employees will flow through to GDP. For example, GDP for 2007 will be revised up about $30 billion, reflecting an upward revision to state and local government spending of about $80 billion that will be partly offset by a downward revision to federal government spending of about $50 billion.

So, for 2007, state and local governments will have contributed an additional $80 billion to the GDP of the United States simply by promising to pay pensions in the future.

There are good arguments for keeping books this way.  For one thing, it will give a better sense of how much of our economy is made up of the government… and the debt required to support it.

On the other hand, I infer from the fact that the explanation makes no mention of discount rates that the BEA is going to accept employers’ calculations of normal cost.  That is, whatever investment return assumptions the state and local governments use to calculate how much they have to pay into their pension systems each year will simply be accepted for the aggregate calculation.  And those assumptions are notoriously optimistic.

That means the real value of each year’s promises is actually much greater than will be captured on paper… and the real amount of debt implicitly being imposed for repayment by future taxpayers is much higher.

The philosophical consequence may be greater, though.  The statistics released by the federal government will make no adjustment for the risk that taxpayers will at some point be unable or unwilling to make good on the promises of prior elected officials.  That makes some sense if you’re thinking within the accountant’s box, but reality is not best viewed through green eye shades.  A promise may be an asset, in some sense, but it’s an abstract asset that doesn’t necessarily correspond with psychological value or physical resources.

As the Popsicle-stick palace of pension promises begins to bear the much-more-real weight of payouts and economic stagnation, the BEA may see its GDP numbers crumble.  Not many people may notice, though, as Americans will be dealing with the consequences of building an assumption-based economy.