An Earmark by Any Other Name Is Still an Earmark

Report Budget and Spending

An Earmark by Any Other Name Is Still an Earmark

February 3, 2009 4 min read Download Report
Karen Campbell
Karen Campbell

An analysis of H.R. 1, also known as "the American Recovery and Reinvestment Act of 2009," passed by the House of Representatives, shows that it fails to deliver on the economic promises of its authors, does not help the economy recover, and--worst of all--decreases investment rather than stimulating reinvestment.

Economists in the Center for Data Analysis at The Heritage Foundation used a widely respected model of the U.S. economy, the IHS Global Insight U.S. Macroeconomic Model, to estimate the economic effects of H.R. 1. Simulation results show:

  • The 10 year average increase in GDP is only $72 billion;
  • The 10-year average number of jobs created is 616,000; and
  • Non-residential investment, which drives sustainable growth, actually decreases by an average $3.1 billion.

Debt should be used only to purchase assets that are income-producing--that is, assets that earn a return greater then the cost of borrowing. If the borrowed money cannot be paid back from the additional income produced by the investment, then the debt becomes a burden on the existing income steam. Many Americans have experienced this economic reality firsthand, and yet the nation's leadership is contemplating assuming this kind of debt--only on a much larger scale.

Borrowing close to $1 trillion to purchase a hodgepodge of appropriation earmarks not only does nothing to stimulate the economy, but it will actually further weaken the U.S. economy.

The Congressional Budget Office estimated the outlays over the next 10 years for each of the various spending and revenue categories. Using these estimates and outlay categories, analysts at The Heritage Foundation simulated the macroeconomic effects of H.R. 1. The simulation used a large structural econometric model of the U.S. economy.[1]

The results of H.R. 1 are compared to the tax simplification and reduction plan proposed by The Heritage Foundation in Table 1.[2] Both plans were simulated using the same model.

The maximum number of jobs created in any one year over the 10-year period from 2009 to 2018 by H.R. 1 is 1.6 million; this maximum does not occur until 2011. The largest increase in GDP comes in 2010 at $173 billion. Thereafter, GDP increases drop off considerably despite the government's continued borrowing and spending. The 10-year Treasury note--one measure of borrowing costs--is on average 0.5 percentage points higher throughout this time period. By 2012, the higher cost of borrowing starts diminishing investments, causing investments to be significantly less than the baseline.

Investments are a key driver of economic growth because they are made in anticipation of the investment improving productivity. The return on investment is the higher income created by the increase in the real quantity of output or the increase in the real quality of output. Higher income allows the investor to pay back the debt plus interest and still have additional income as a reward for taking the investment risk and growing the economy.

Since the government's revenue comes from taxes on the income produced in the private sector, government borrowing relies on the underlying productivity of the national economy in order to repay the principal plus interest on its debt. The lower level of new investments weakens the productivity of the economy making the H.R. 1 spending spree even more burdensome as interest and principal on this debt comes due.

A National Burden

H.R. 1 does not stimulate the economy. Indeed, by the CBO's estimates, most of the spending will occur well after the economy would have recovered on its own.

This government consumption spending creates a debt burden as lower investment spending diminishes the income streams available to pay the principle and interest on the debt. The debt will either have to be repaid with existing income streams or rolled over at higher interest rates. Consequently, investors will have to forego even more productivity-producing investments.

Rather than helping the economy recover, this bill further weakens the economy.

Appendix: Methodology

Analysts used the IHS/Global Insight model[3] of the U.S. economy and their October baseline[4] to simulate the effects of H.R. 1. The assumptions were based on the "Congressional Budget Office Cost Estimate: Bill HR-1: The American Recovery and Reinvestment Act" of January 30, 2009.

The spending categories[5] were placed into one of five aggregate variables used by the model: government non-defense consumption, government defense consumption, non-Medicare transfers to state and local government, Medicare transfers to state and local government, and cyclical federal transfers to individuals. These variables were increased by the amounts and years[6] specified by the CBO report.

The revenue effects were estimated by their effect on the average effective tax rate. This was calculated by reducing the "personal income tax receipts" variable in the model by the amount of the CBO-estimated revenue loss. The ratio of the effective tax rate to the previous tax receipts was used to estimate the H.R. 1 effective tax rate. The tax credits lowered the average effective tax rate by approximately 1-3 percent per quarter.

Lastly, the variable estimating the federal debt held by the public was increased by the amount of the deficit created in each year as estimated by the CBO report. The model was then run and the results obtained.[7]

Karen A. Campbell, Ph.D., is a Policy Analyst in Macroeconomics in the Center for Data Analysis at The Heritage Foundation.

[1]See the appendix for details.

[2]J. D. Foster and William W. Beach, "Economic Recovery: How Best to End the Recession," Heritage Foundation WebMemo No. 2191, January 26, 2009, at

[3]For information about the details and operation of the Global Insight U.S. Macroeconomic Model, see "Description of the Global Insight Short-Term U.S. Macroeconomic Model," at /static/reportimages/53A9877A60ECA7BEF5E99A1744A27AA5.pdf .

[4]The October baseline was chosen because baselines after this month began including stimulus spending by the federal government. Therefore, some effects would already be built into the baseline, which would have caused a type of double counting.

[5]The only "spending" that was not included in the model is the direct spending provision for health information technology. It was unclear how this provision would be implemented. The total amount of this appropriation is $20.2 billion. Its effect in the model would most likely be negligible.

[6]The CBO report estimated outlays in 2019. The baseline used for the simulation went to 2018 only. The additional spending in 2019 was $3.3 million. This would have an insignificant effect on the model results.

[7]Detailed results and assumption calculations are available upon request.


Karen Campbell
Karen Campbell


Religious Liberty

The Natural Law Foundation of Freedom of Thought, Conscience and Religion

Watch live now