LBS logo London experience. World impact.

The relationship between US stocks and government

There’s a new stock selection strategy in town that has generated abnormal annual returns of up to 6.9 per cent.

By Vito D Gala Jun Li and Frederico Belo . 25 July 2013

The-relationship-between-US-stocks-and-government-974x296

Forget bottom-up or top-down stock picking for a moment. Stop poring over those share charts. Set aside notions of technical or fundamental analysis. There’s a new stock selection strategy in town, backed up by some impressive research, that has generated abnormal annual returns of up to 6.9 per cent.


In their new paper, academics Frederico Belo (University of Minnesota), Vito Gala (London Business School) and Jun Li (University of Texas), outline the relationship between the performance of certain US stocks and the political party occupying the Oval Office.


Based on the premise that government expenditure is greater during Democratic presidential cycles, the paper focuses on government spending and its effect on stock prices. Using detailed industry-level data from the US National Income and Product Accounts (NIPA) input-output accounts, the authors devised a measure of industry exposure to government spending. In this way they were able to tell which industries have a higher proportion of their total output purchased directly by the government sector, as well as indirectly through the chain of economic links across industries.


Industries with the highest exposure to government spending included aircraft, defence and construction, as well as secondary industries such as steel and mining. While industries with some of the lowest exposures included tobacco, alcohol and food products.


Next, five stock portfolios were constructed, representing different exposures to government expenditure. The performance of these portfolios was tracked across the period 1955 to 2009, through the tenures of Democratic and Republican presidents. To make the research more robust, portfolio performance was also examined from 1929 to 2009.


At the same time, the authors made sure that any pattern in the returns was not due to a range of other possible factors including: business cycle effects; firms’ characteristics including size; book-to-market ratio; momentum; market beta, or corporate political contributions.


The research results are something of an eye opener. Especially for anyone who has spent hours sifting through corporate data in an attempt to separate winners from losers, and identify potential investment targets.


The study suggests that, depending on which party has its representative in the White House, it is possible to predict the profitability of a cross section of firms, according to their exposure to government spending. When there’s a Democratic president, firms with a high government spending exposure are more profitable than those with low government spending exposure, all things being equal. The reverse is true for Republican presidents.


Correspondingly, the stock returns for a portfolio of firms with a high exposure to government spending are greater during a Democratic presidency. While during a Republican presidency returns for firms least exposed to government spending fare best. It is a surprising pattern, especially given the difference in spending between presidents from the same party. Yet it was true for all partisan presidential periods, bar three — the Franklin D Roosevelt term in 1937-1941, and the John F Kennedy and Lyndon B Johnson years of 1961-1965.


Returns were impressive, too. An investment strategy exploiting the presidential political cycle generated abnormal excess returns up to 6.9 per cent per annum. These abnormal returns were usually obtained in the second and third years of a presidential term, when there was greater certainty about government spending policy.


So, if you are short in stocks exposed to government spending, maybe now is the time for a rethink.


This is an executive summary of the latest research by Frederico Belo, Jun Li and Vito Gala. 'Government Spending, Political Cycles and the Cross Section of Stock Returns.' Journal of Financial Economics.

Comments (0)

lbsr-thedonald-subscribe-promo-desktop

Subscribe to LBSR

Get the latest ideas and opinion from London Business School’s experts, straight to your inbox