10/16/2009 05:12 am ET Updated May 25, 2011

Irresponsibly Following the Congressional Budget Office

Suppose that a prominent member of Congress were to press for spending an additional $450 billion a year over the next three years on infrastructure projects. Suppose that they did this at a time when the unemployment rate was near 4.5 percent, the definition of full employment used by the Congressional Budget Office (CBO) and in most econometric models.

The CBO modeling of this proposal would show that infrastructure spending would lead to almost no increase in employment. It would show that the effect of the additional spending would be higher interest rates, which would crowd out investment and other forms of expenditures. The spending would also raise the deficit and lead to larger interest rate burdens on our children.

This prominent member of Congress would undoubtedly be denounced as irresponsible on the editorial pages of the Washington Post and by other respectable pillars of the Washington establishment. However, if it was the winter of 2007, this prominent member of Congress would have been performing an extremely valuable public service, especially if she was able to push her plan through Congress.

The reality is that the U.S. economy was about to fall off a cliff in the winter of 2007, but CBO and most other economic forecasters were completely oblivious to the problems that stood in front of their face. They either did not see the housing bubble that was collapsing around them or somehow thought that its collapse would not have any major impact on the economy. As we all now know, CBO was hugely wrong.

This is not the only time that CBO has been wrong in a really big way. They often miss major economic turning points. For example, in January 2001, the projections showed a respectable 2.4 growth rate for 2001 and a 3.4 percent growth rate for 2002. The unemployment rate was projected to average 4.4 percent in 2001 and 4.5 percent in 2002. There was no hint of the recession that began just two months later.

CBO did not even recognize that the collapse of the stock market bubble, which was already in progress, would affect capital gains tax revenue. This caused it to overestimate capital gains tax revenue by more than $300 billion over its 10-year projection period, an amount equal to approximately 1 percent of the budget and 10 percent of the deficit.

It's not just economic turning points that CBO tends to miss. They sometimes badly misjudge the impact of specific programs. Last summer Congress funded a program to keep homeowners facing foreclosure in their homes. CBO projected that 400,000 loans would be modified under this program by 2011. Through April of 2009 there had been fewer than 1000 applications, and only 51 completed modifications.

A column in the New York Times last week noted how CBO had repeatedly underestimated the cost savings that resulted from various efforts to restrain medical care costs. Jon Gabel, a researcher at the University of Chicago, argued that CBO assumes zero cost-savings for any legislated changes where the benefits are not known.

As Gable argues, the issue is not that the CBO is partisan or biased in any obvious way; there is no basis for questioning its integrity. Nor is it a lack competence; CBO is staffed by hard-working professionals.

However, they are nonetheless fallible. CBO does make mistakes, and often they are large and important mistakes. The failure to recognize the housing bubble, and that its collapse would have a devastating impact on the economy (and the budget deficit) was an enormous error. The economists at CBO may take solace in the fact that most of the economics profession made the same mistake, but this fact provides little consolation to the tens of millions of workers who are unemployed or underemployed as a result of the failure of CBO and others to warn of this impending disaster.

The fact that CBO is fallible is important for how members of Congress view its analysis and projections. Members of Congress would be foolish to ignore the projections provided by CBO, however they would be irresponsible if they treated this analysis as the final word.

Members of Congress are responsible for getting the policy right, not doing what CBO tells them. This means that if they have reason to believe that a CBO projection in a specific area is wrong, they should act based on their judgment, not on the CBO projection.

Constituents have every right to hold their representatives in contempt if they try to blame their mistaken judgments on CBO projections. Members of Congress get paid for getting the policy right, not listening to CBO.

To take the specific example mentioned at the start of this column, members of Congress should have been pushing large infrastructure projects at the beginning of 2007. These projects would have kicked in just when the collapsing bubble was sucking the life out of the economy.

Pushing for major infrastructure spending at the start of 2007 would have caused members of Congress to be denounced as irresponsible by the Washington Post and the rest of the Washington establishment. But it is much better to be denounced as irresponsible than to actually be so -- and blindly listening to CBO is incredibly irresponsible.