US Defence Contractor Stocks Provide Evidence Of The Impact Of Government Spending Shocks

How does the economy respond to a surprise change in government spending? A study by Jonas Fisher and Ryan Peters, published in the May 2010 issue of the Economic Journal, uses 50 years of data on the stock returns of US military contractors to identify government spending shocks and their impact.

The main results are based on the stock returns of military contractors that have ever appeared in an annual listing of the top three US military contractors by value of primary contracts awarded from 1958 to 2007.

The total sales of the top three contractors follow the contours of aggregate US defence spending, and excess returns of these firms have superior explanatory power for defence spending and government spending as a whole.

The researchers find that after a positive excess return surprise, output, hours and consumption initially are flat for several quarters, and then all rise in a hump-shaped pattern.

Real wages initially decline during the period when the other variables are flat. After this initial decline, wages rise persistently over the period when output, hours and consumption are following their hump-shaped paths.

Finally, the study finds that the government spending output multiplier is about 1.5 over a horizon of five years.

The question of how the economy responds to a surprise change in government spending is clearly important from a public policy perspective, particularly given current events. The two main approaches to identifying government spending shocks lead to different answers and they both have important shortcomings.

This study develops and applies a new measure of government spending shocks based on defence contractor stock returns, which avoids the shortcomings of these approaches.

The narrative approach identifies specific dates when persistent increases in US military spending were anticipated and uses them to identify government spending shocks.

The principal advantages of this approach are that the shocks are genuine surprises and that it takes into account the anticipated nature of the government spending. Shortcomings include the small number of episodes, the episodes only involve spending increases, the need to assume that the spending is certain, and the approach''''s inherent subjectivity.

The innovations approach identifies spending shocks with the statistical surprises to government spending in a dynamic regression framework.

The principal advantages include that there are many ''''dates'''', both positive and negative shocks, and it is less subjective. But there is one key shortcoming. Government spending is often anticipated before it is recorded in the data and failure to take this into account may lead to biased estimates.

Like the narrative approach, the starting premise of Fisher and Peters'''' approach is that there exist surprise shocks to military spending. Given such a shock, current and expected earnings of military contractors change. Forward-looking agents incorporate these expectations into their stock valuations, thereby affecting the returns to holding these stocks.

Not all variation in contractor stock returns is due to expectations about military spending. The researchers address this in several ways:

  • First, they identify spending surprises with surprises in military contractor stock returns in excess of the market and uncorrelated with the state of the economy.
  • Second, they combine stock returns of multiple contractors.
  • Third, they emphasise the returns of large contractors that specialise in military production.

Their approach increases the number of ''''dates'''', includes positive and negative shocks, is less subjective than the narrative approach, is immune to the anticipation critique, and incorporates the inherent uncertainty involved in predicting the future path of defence spending.

''Using Stock Returns to Identify Government Spending Shocks''' by Jonas Fisher and Ryan Peters is published in the May 2010 issue of the Economic Journal.