CBO Stimulus Estimates

Totally agree. Plus, one has to ask: what kind of output? a GDP number is not an end in itself.

Doug also reports:

"CBO estimates that by 2019 the Senate legislation would reduce GDP by 0.1 percent to 0.3 percent on net."

But, there are very few data points.

Not being a macro guy, I'm not sure how to evaluate these figures. As a micro guy, numbers to the second decimal place leave me highly skeptical. What's their track record in predicting GDP?

As a scientist, I'd like to know if there is a suitable control estimate, i.e., what would growth be absent any 'stimulus'? Would it be 1.1 percent to 5.4 percent? Even if there were an incremental benefit to stimulus, would it be worth an extra trillion $ (Don is right; GDP isn't the whole game)?

Even if the aggregate GDP growth is provably, incrementally better, is that worth the massive exercise in state power that will eventually be needed to repay those sums? From our kids and grandkids?

Besides my epistemological, economic, and moral concerns, I think this sounds just great.

What you are failing to account for is how much the government will have to borrow even if it is not passed. Just because there is no stimulus package does not mean there will be no increase in borrowing. There will very likely be even more. This could quite easily pay for itself.

Could you explain what you mean by "this could quite easily pay for itself"?

The CBO does use crowding out effects and assumes a temporary zero net effect stimulus, mainly moving consumption forward. The variance in the estimates seem to correspond in magnitude to the variance in deflationary pressure as seen in Krugman's chart on price vs growth.

That variance is based on the uncertainty of goods elasticity, mainly energy and commodities. If critical supplies suddenly become short, then the whole stimulus result will move forward in time, more rapid rise, then a more rapid fall in GDP.

In other words, if we get the critical elasticities wrong then we get a double dip recession.

MHodak: "What's their track record in predicting GDP?"

That's an excellent point. Very few models can predict the next quarter of GDP with reasonable accuracy, let alone GDP for the next 2 years. But even more important, GDP increases do not mean increases in wealth. Because of the way GDP is constructed it is primarily a measure of consumption. GDP increased dramatically during WWII, as an extreme example, but wealth did not. GDP increased then because of state spending on war materials. Only spending on capital goods, that is, investment, can make us wealthier.

I'm not clear why a world distaste for US government debt would be the end of the world. It seems to me like the only thing that has any chance of giving us a balanced budget.

The effect of a change in inflation from 3% to -3% on the real debt is $800B alone. The effect of a longer deeper recession could easily mean a similar amount in reduced revenue and increased automatic spending (unemployment, etc.), both of which are the real drivers of debt.