International Evidence on Firm Level Decisions in Response to the Crisis: Shareholders vs. Other Stakeholders
One of the interesting features of the 2008 financial crisis is the wide range of relationships between changes in a country’s output and changes in unemployment. Spain and Ireland had very large increases in unemployment despite quite different falls in output.This is perhaps not very surprising because both had significant construction industries that were devastated by the bursting of the property bubbles in both countries. More surprising is the fact that countries like Germany and Japan had much larger drops in output than the US but the effect on their unemployment rates was small.
The path of gross domestic product (GDP) for the G5 countries, France, Germany, Japan, the US and UK between quarter 1, 2008 and quarter 1, 2009 involved wide variations.Japan had a 10% drop, while Germany’s dropped more than 6%, for the UK and US it was about 5%, and for France under 4%.However, for unemployment the paths were dramatically different from these figures.Japan’s unemployment increased a small amount, while Germany’s fell most of the time. France and the UK rose somewhat but by far the largest change was the US, which more than doubled from just under 5% to 10%.
In a well-known paper from 1962, Okun used data from quarter 2, 1947 until quarter 4, 1960, and found that a 3% change in GDP was associated with a change in unemployment of about 1%. This relationship became known as Okun’s law.Although it was recognized that Okun’s law varied across countries and time, the breakdown of the relationship during the crisis was of a different order of magnitude than what was previously observed.
There has been extensive discussion of why this change has occurred.One important point that is frequently made is that Okun’s law is a statistical relationship. It is not based on a theoretical framework.There is no particular reason why the relationship should be expected to be stable. The examples of Ireland and Spain suggest that the importance of the construction industry in employment when there is a real estate bubble that bursts is a key factor. Much of the discussion has been focused on differences in labor markets. In particular, there are significant differences in employment protection law, the share of temporary workers not protected from dismissal and the generosity of unemployment insurance. While there is no consensus on the importance of these factors, some authors conclude that it is difficult to identify a robust relationship between cross-country estimates of Okun’s coefficient and labor market institutions.
One important institutional difference between the countries that has not been considered in the previous literature concerns corporate governance. In the UK and US it is quite clear that shareholders own the firm and managers have a fiduciary (i.e., very strong) duty to act in their interests. In contrast, in Germany there is co-determination. In large corporations employees and shareholders have an equal number of seats on the supervisory board of the company. Here workers' interests will also matter. In Japan, managers do not have a fiduciary responsibility to shareholders. The legal obligation of directors is such that they may be liable for gross negligence in the performance of their duties, including the duty to supervise. In practice, it is widely accepted that stakeholder interests and in particular employee interests play a predominant role. The system in France is that partially privatized companies must reserve two or three board positions (depending on board size) to be elected by employees. Also, employees in companies where at least 3% of shares are employee owned have the right to elect one director.
Governance differences among firms across countries appear not only in written laws but also in management and firms' approaches. For example, survey results of Japanese firms, show that the majority of managers consider employees and customers as the most important stakeholders. Only a minority of them consider investors as highly important.
Our paper takes a different approach than the existing literature in that we consider the G5 countries and focus on the question of how firms in the different countries reacted to the shock that the financial crisis inflicted. While adjustments in labor are no doubt important, they represent just one margin. There are many others, including investment, financial structure, payout policy and so forth. Our paper considers a whole range of adjustments that firms can make and that can represent important differences in the way firms react in times of crises.
To analyze firms’ response to the crisis, we consider firm level data rather than aggregate data. We take firm level accounting data from Worldscope for France, Germany, Japan and UK and Compustat for the US. We focus on the largest 20% of firms in these datasets. These account for much of each country’s economic activity in each of the G5 economies.
One important difference across countries is that the major industries vary considerably in size. For example the automobile industry is very important in Germany but not in the UK. To deal with this we match firms in France, Germany, Japan and the UK with similar firms by size and industry in the US. We then consider how these firms in different countries reacted to the shock of the crisis.
We find significant differences between the response of US and non-US firms. US firms significantly decreased their production costs relative to firms in other countries. They also reduced debt and dividend payout, and increased their cash holdings compared to firms in other countries. We find that the differences are, in general, explained by differences in financial leverage between US firms and foreign firms. Higher financial leverage in US firms before the crisis made firms more vulnerable to funding difficulties in the financial markets and led to more drastic changes in their production decisions. We argue that financial leverage does not explain differences between production decisions in German and US firms and between Japanese and US firms. Rather, we find evidence consistent with the hypothesis that differences in firm governance between US firms and firms in Germany and Japan drive these responses. In particular, US firms are more prone to cut labor costs and reduce leverage compared to German firms and Japanese firms in order to achieve larger profits and a larger cash-cushion in the short-run.