15 May 2019

Putting workers on UK boards wouldn’t be as bad as its critics fear, or as good as its advocates hope. But on balance the evidence doesn’t support compulsion.

The UK Corporate Governance Code says that worker directors are one of three options for ensuring employee voice is heard in the boardroom. But the reaction to Capita’s recent announcement that it has appointed two employees to the board just goes to emphasise how few companies are introducing this model by choice. Indeed evidence from around the world is that employee directors are hardly ever appointed when firms are not compelled to do so. Some academics have taken this as evidence that worker directors can’t be good for companies (or at least for their shareholders), otherwise we’d see them voluntarily adopted more often.

This lack of voluntary action is leading to calls to make board representation of workers compulsory. In 2016, UK Prime Minister Theresa May dipped her toe in the water before finding it to be too hot. The Labour Party has made it a core policy proposal, saying it will mandate companies with 250 or more employees to have one-third worker directors. Even in the US, Democratic Senator Elizabeth Warren has made worker directors a centre-piece of her Accountable Capitalism Act. Putting workers on boards is gaining favour as a way to rebuild broken trust in capitalism.  But is there any evidence that it would make a difference?

This question was discussed recently at a conference on ‘Who should run a company?’ organised by London Business School’s Centre for Corporate Governance. Addressing the conference, Dr Ernst Maug, Chair of Corporate Finance at the University of Mannheim, presented an illuminating overview of research on the German experience, including his own latest findings. A summary of his research and link to the full paper can be found here.

German regulations provide for tiered levels of board representation depending on number of employees. Firms with more than 500 employees must have one-third of of directors elected by employees, rising to one-half of directors where there are more than 2,000 employees. Maug's research, carried out with Christoph Schneider and E. Han Kim, looked at differences in wages and employment between German ‘parity firms’ (those with half of Supervisory Board seats allocated to worker representatives, with the Chair having the deciding vote in the event of a tie) and 'non-parity firms' (those firms with one-third of directors or fewer elected by employees). What they found was that parity representation created the conditions for firms and employees to enter into an implied employment insurance contract: workers earned 3.3% less on average, but in exchange were protected from layoffs in the event of negative industry shocks. Employees traded lower wages for higher job security. This effect was not observed in firms with one-third representation, suggesting that workers and management can only strike the implied bargain when the institutional set up gives some assurance that it will be honoured.

Interestingly the protection only extended to white collar and skilled blue collar workers. Unskilled workers experienced layoffs at the same level in parity and non-parity firms. The researchers hypothesise that this is because the unskilled workers were not effectively represented by the elected directors. They found that the worker board members were drawn almost exclusively from white and skilled blue collar backgrounds, and were frequently ‘professional’ worker directors provided by the unions rather than drawn directly from the company workforce. 

Turning to the impact on performance, the research found no positive or negative relationship between parity representation and either firm profitability or share prices. This is at odds with other evidence, which, while very mixed, tends to find German co-determination to be positive for productivity and profits but negative for share valuations.

How could worker directors be good for productivity but bad for shareholders? One explanation is that worker directors may contribute to: more co-operative labour relations; encouragement for employees to invest in building firm-specific skills and human capital; and, as found by Maug and co-workers, acceptance of lower wages in exchange for job security. In the normal course of business, this leads to higher productivity. But it comes at a cost. In industry shocks or periods of rapid change, co-determination may make it more difficult for firms to cut costs, or to extract capital and redeploy it to more productive firms or industries. Management and employees may get into a mutually entrenched position whereby each guarantees the other's job security. There is evidence that parity board representation at firms may cause shareholders to take offsetting action to mitigate the risk of entrenchment, for example by increasing leverage above otherwise optimal levels. Shareholders may view the cost of this loss of long-term optionality as outweighing any short-term productivity benefits.

Ernst Maug’s presentation left me drawing a number of conclusions:

1. Overall the evidence on the impact of worker directors is mixed, suggesting it is neither a terribly good nor a terribly bad idea. Mandating workers on boards would not be the end of Capitalism As We Know It, but nor would it lead to a nirvana of higher economic growth and trustworthy corporate behaviour. The conclusion you draw from this will depend on your political persuasion: either you’ll conclude that because it’s not so harmful, we should try it to rebuild trust; or you’ll conclude that there’s insufficient evidence to force upon shareholders a governance model that they clearly don’t willingly choose.

2. The institutional details of worker directors matter greatly. In Maug’s research, the major effects, both positive and negative, appear in the German system when parity is reached on the Supervisory Board. Moreover, the selection and background of worker directors has a critical effect on who benefits from their representation. Arguably in the German system, worker directors may represent the interests of union members as much as the interests of employees in the firm. Worker directors can therefore create problems of insiders and outsiders, which may pit represented against non-represented employees, or current versus potential employees. Any eligibility and selection process needs to be very carefully thought through.

3. Current employees do seem to benefit from parity representation in the German system, through an implicit insurance contract, which improves job security, albeit at the cost of a reduction in wages. Overall welfare for existing employees is likely to be increased.

4. Evidence on the effect workers on boards have on companies is mixed, but overall suggests that the impact on productivity and profits is neutral to positive, but the impact on shareholders is neutral to negative. This suggests that shareholders incur costs that are not reflected in near-term productivity metrics (which might be, for example loss of optionality in times of change or shock).

5. Introduction of employee directors should be counterbalanced by strong shareholder rights to avoid mutual entrenchment of management and employees. Recent corporate governance scandals in Germany may come to be seen as case studies of the deleterious effects of combining weak institutional investor oversight with an entrenched symbiosis between management and employee representatives.

6. The economy-wide effects of worker directors may be more damaging than the firm-specific studies suggest. Any benefits from productivity or labour relations at the firm level could be outweighed by the economy-wide costs of inflexibility of labour or capital deployment in times of shock or change. Worker directors may entrench the status quo and inhibit reallocation of resources from lower to higher growth industries. This would harm economic growth and employment overall, and would benefit workers in firms with employee representation at the expense of the unemployed, non-represented workers, and shareholders. Research on the broader economic implications of worker directors remains very limited, in part because of the difficulty of untangling the various factors in play.

There’s enough conflicting evidence out there for people to pick and choose the studies that fit their prior convictions. But the fact that the evidence does not all point one way suggests that real care is taken in this area and that the details of any proposal matter greatly. The careful work of Maug and his co-authors shows where some of the trade-offs may lie, trade-offs that are ultimately a matter of political choice. 

For what it’s worth, my own view is that the evidence doesn’t support mandatory adoption of workers on boards. Evidence abounds that companies should be focussed on obtaining and acting on meaningful employee insight, every bit as much as they are for customers. But this doesn’t require worker directors. And as we enter a time of profound economic change, inhibiting the process of reallocation of capital from old to new industries could be immensely damaging for the economy in the long term. Worker directors risk entrenching the status quo and undermining the process of creative destruction that is so essential to growth. The evidence of the benefits of worker directors is not nearly compelling enough in my view to both offset this risk and to justify imposing on shareholders a governance model that they clearly don’t want. 

I recognise that others will look at the same evidence and take a different view. But I hope at least that a careful review of evidence will guide people to where the risks lie. Moreover, good companies must still work hard to ensure authentic and meaningful dialogue with employees about matters that affect them. The UK Corporate Governance Code lays down a challenge to companies to achieve this, which companies must respond to, or they'll have no reason to complain if worker directors are forced upon them.