A prize-winning study into partisanship shows it is no myth business performance is affected by ruling party orientation

By Sasha Molchanov, Art Durnev and Jon Garfinkel*

Overall, it’s difficult to draw sweeping conclusions from the literature regarding the influence of partisanship on financial or economic outcomes.

Depending on the data analysed, stock returns have been found to be higher under either Democratic or Republican presidencies in the US, and another study² finds no evidence of a presidential cycle in US industry returns.

The lack of consensus is particularly concerning when economics plays a critical role in the outcomes of elections.

Economic issues often take central stage during electoral campaigns, with ‘leftist’ parties frequently proposing policies that are very different from those of ‘rightist’ parties, particularly in their expected impact on the business environment.

Such policies include stringent labour and environmental laws, higher taxes, and policies that may encourage higher interest rates (which increase the cost of borrowing). Thus, understanding the link between partisanship and economic outcomes is important from a policy perspective.

Finessing the frameworks

Three factors that appear to be overlooked in prior studies may explain the puzzling lack of a consistent relationship between party-in-power orientation and corporate performance.

Factor 1:
Not all firms are equally sensitive to government policies. For example, while labour-intensive firms are likely to be adversely affected by stringent labour laws, capital-intensive firms are less likely to be.

Factor 2:
The link between ruling party orientation and leftist legislation is sometimes weak (this is documented in the literature and confirmed in our analysis). For example, legislation often regarded as ‘leftist’ is some-times enacted under ‘rightist’ governments.

Factor 3:
Much of the existing literature uses a simple dummy variable approach, unambiguously classifying all governments as either ‘left’ or ‘right’ and thus ignoring potentially important characteristics (such as coalition governments, or legislative and executive branches being controlled by different parties).

In order to address Factor 1, we posit that leftist legislation has four dimensions and that not all firms are equally affected by such legislation.

First, if leftist legislation is more ‘friendly’ to labour, then we would expect more-labour-intensive industries to have greater sensitivity to labour-friendly legislation and perform worse under it.

Second, leftist environmental legislation is more likely to be stringent, adversely affecting highly polluting firms.

Third, leftist tax policy favours higher rates, and so we would expect firms with higher gross profit margins to experience higher tax bills, although their overall performance (stock returns) may be better.

Finally, leftist governments may enact policies that are traditionally associated with higher interest rates. Thus, we would expect firms more affected by the cost of borrowing (that is, firms with high leverage) to experience drops in performance when interest rates are increased.

To address Factor 2, we note there is ample evidence of legislation that is traditionally viewed as leftist actually being passed when rightist governments are in power. Such imperfect correlation between a ruling party’s orientation and the policies it implements may be yet another reason for the lack of consistent relationship observed between government partisanship and corporate performance.

We address this by analysing only those policies that are explicitly associated with a government’s party orientation.

Finally, most extant research on political cycles in economics and finance relies on a simple dummy variable approach to indicate a ruling party’s orientation (we have noted this above as Factor 3). While intuitive, such methodology ignores more-complex government arrangements that occur – such as when no party has an outright majority or when no single party controls all government branches. For example, a dummy variable approach would classify ruling-party orientation in the US as ‘leftist’ both before and after the 2010 Congressional (Senate and House of Representatives) elections, even though the Democrats lost control of the House of Representatives after the election.

To address this, we developed a unique five-point scale. A score of five corresponds to ‘left-wing dominance’ – that is, a leftist party controls both the executive and legislative branches of government (in a parliamentary system, the leftist party would control more than two-thirds of the Parliament). A score of one would represent the opposite – a rightist party in control. A score of four (or two) corresponds to a lower degree of control by the leftist (or rightist) party; and a score of three represents a centrist government.

In this classification, the current New Zealand government would receive a score of two. The National Party (classified as rightist) leads the ruling coalition; this coalition, however, holds less than two-thirds of the seats in Parliament.

Divining the data

To conduct our tests, we built a sample at the industry level using 57 industries from 50 countries during the years from 1990 through to 2006. We constructed four industry-level sensitivities to leftist legislation.

The first sensitivity measure was labour intensity.

Leftist governments are often thought to be associated with strict labour legislation, which increases the costs of operation and makes labour-capital substitution more difficult. We measured labour intensity as the ratio of the value of labour inputs to the total value of production inputs.

Our results were consistent with our expectations: labour-intensive industries had lower valuations, lower returns on assets (ROAs), and lower stock returns when leftist governments were in power. As for the actual policy measure (rigidity of employment legislation), we confirmed that labour laws are stricter under left governments and that employment-legislation rigidity explained by ruling-party orientation³ has a significantly negative impact on labour-intensive industries’ performance measures.

Our second sensitivity measure was environmental legislation.

Leftist governments are often linked to a tightening of environmental standards. Thus, firms that are less environmentally friendly will be adversely affected by the enactment of such legislation.

To compute sensitivity to environmental legislation, we used an index of environmental responsibility based on rankings obtained from MSCI’s ESG (environmental, social and governance) database ratings.

We confirmed that industries which are more environmentally sensitive experienced worse performance (in terms of ROA, valuation, and stock returns) when leftist parties were in power. When we used the actual policy measure (rigidity of environmental legislation) explained by ruling-party orientation, the results were similar: environmentally sensitive industries suffer a drop in performance when ‘leftist’ parties enact tighter environmental standards.

Our third sensitivity measure was the corporate tax rate.

Parties on the left of the political spectrum are often viewed as supportive of higher corporate tax rates, which decrease after-tax income and may discourage entrepreneurial activity. We measured tax rate sensitivity using gross profit margin (ratio of EBIT to sales) and found that tax-sensitive industries have lower ROAs and stock returns under leftist governments.

The actual tax rate we used was a five-year effective rate. Even though the relationship between this tax rate and the ruling-party orientation is not statistically significant, the portion of tax rate explained by ruling-party orientation, however small it is, has a significantly negative impact on the performance of more-tax-sensitive industries. This implies that even though leftist governments do not always raise taxes, when they do, the impact is significant.

Finally, we considered sensitivity to interest rates.

Prior research has documented higher interest rates under leftist governments. Because this increases the cost of borrowing, firms with high leverage (a high ratio of total debt to assets) should be more adversely affected by the higher interest rates associated with leftist govern-ments. Our results confirmed our expectations: industries more exposed to interest rate movements have lower returns and value when leftist governments are in power.

Interest rates explained by ruling-party orientation also have a significantly negative impact on performance of interest-rate-sensitive industries.

Confirming causality

Analysis of the impact of political environment on financial outcomes is hampered by potential reverse causality. Our results established that a government’s political orientation has a significant impact on the performance of policy-sensitive firms.

However, economic performance has a potentially strong impact on electoral outcomes.

Our sensitivity approach analyses industry performance within each country and it is less likely that within-country differences in performance have a strong systematic impact on political variables.

Nevertheless, we explicitly addressed potential reverse causality in a number of ways. First, we controlled for past economic performance by including a number of lagged economic variables. Second, because electoral rules in most parliamentary systems allow for ‘called’ elections and hence make strategic electoral timing possible, we performed the analysis on sub-samples of presidential and parliamentary systems. Third, we performed a two-stage regression, obtaining a party-orientation index explained by past economic performance.

Our results remain robust in all of the above-mentioned tests.

---------------------------------------------------------------------------

This article is based on A Durnev, J Garfinkel & A Molchanov (2012) Partisanship and Corporate Performance, which won the ISCR prize for the best paper on financial regulation at the 2012 New Zealand Finance Colloquium. It is available at www.nzfc.ac.nz/archives/2012/papers/updated/39.pdf.

² B Jacobsen & J Stangl (2007) ‘Political cycles in industry returns’ Journal of International Finance and Economics 1 pp113-130.

³ To establish this, we performed a two-stage estimation. In the first stage, measures of ‘leftist’ policies (labour and environmental legislation strictness, corporate tax rates and interest rates) were regressed on the measure of leftist-party orientation. In the second stage, we regressed firm performance measures on ‘leftist’ legislation explained by ruling-party orientation.

---------------------------------------------------------------------------

*Alexander (Sasha) Molchanov is a senior lecturer at Massey University’s School of Economics and Finance in Auckland. Art Durnev and Jon Garfinkel are respectively assistant professor and associate professor of finance at the University of Iowa’s Henry B. Tippie College of Business.

This article was first published by the ISCR in the July 2012 issue of their "Competition and Regulation Times". It is republished here with permission.

The ISCR website is here » and a .pdf version of their Issue #38 is here »