Last week I set out the case for greater rather than less involvement of unions in superannuation. It is not that unions are without fault, but rather that there may be a better alignment of interests between unions and investors than between independent trustees and investors, the main alternative that is on offer. If so, governance based on greater involvement of unions should result in better outcomes for investors; for which there is some empirical evidence. This week, I elaborate on a problem that I glossed over last week, information asymmetry and the principal/agent problem.

I start with the premise that there are insiders, those who work in the funds and understand the investments that the funds make; and outsiders, namely investors. I assume that insiders can determine whether the investments are meeting the interests of investors, and outsiders cannot do so (or cannot do so to the same degree). This is known as information asymmetry.

Ideally, we would like outsiders to know as much as insiders so that they can do one of two things: change the amount of money they invest in the fund including switching to other funds if performance is inadequate; or find avenues for improving the funds, perhaps by selecting insiders (i.e. trustees) that better meet their needs. We would like to remove this information asymmetry but that is likely to be costly. Money spent on reducing information asymmetry could, perhaps, be better used to make additional investments.

The insiders are also Agents acting on behalf of the investors or Principals. If the interests of the Agents are perfectly aligned with investors — say both want to maximize returns over 40 years for a given level of risk — then we don’t need to spend investors’ money on reducing information asymmetry. We can be confident that the insiders will invest to meet outsiders’ needs. But if the interests of Agents are misaligned then this is not so.

As I noted last week, the practice of linking executive pay to share price aims to align executives’ interests with shareholders’ interests. Then, it is said, executives will always act in shareholder’s best interests. But the typical share scheme (and tenure of an executive) is much shorter than the life of the firm or the life of investments at the firm. The executives also have much better information on firm performance than shareholders or even board members. As a result, ‘although stock-based compensation induces managers to exert costly effort to increase their firms’ investment opportunities, it also supplies incentives for suboptimal investment policies designed to hide bad news about the firm’s long-term growth.’[1] It presents incentives for executives to maximise performance over their tenure and over the term of their stock options, even if this doesn’t maximise long-term performance.

This problem suggests that there is little chance, given the much longer horizons, of aligning incentives in superannuation using a similar approach. Even independent trustees have a tenure much shorter than the investment horizon of most fund members; it is hard to escape the view they will be most closely focused on performance in the 3 to 5 years that they work at a fund. We might have to rely on more regulation and more unions after all.

But a look back at history suggests that this may be too pessimistic. Stock-based performance schemes are not just a feature of today’s capitalism. Du Pont introduced an executive incentive plan 90 years ago, in 1927, shortly before the first Wall Street crash. Under their “Executive Trusts Plan” Du Pont did not simply gift shares or options to its executives, but rather ‘the company [received] in payment interest-bearing notes running from seven to ten years, with the stock so purchased deposited as security for the payment of the notes…’[2] What did this mean? Basically, the executive incurred a long-dated debt (which Du Pont would lend them) to fund the acquisition of the stock, interest on which compounded (or was paid down with dividends). This gave executives a strong incentive to maximise the value of the stock over the lifetime of the debt.[3]

So, executives had their own skin in the game, in the form of an obligation to repay a substantial debt, and a long-term incentive consistent with the tenor of the debt. Unlike the modern executive, they could not escape the debt by simply walking away.

Could you adapt this to superannuation? Senior executives and trustees could be encouraged to ‘buy’ shares in the funds they manage using, perhaps, market-rate compounding loans provided by the fund maturing in, say, 20 or even 30 years. While free to move on, executives and trustees would be precluded from switching these shares to other funds, or precluded from doing so within, say, 15 years. This, alone, would help to align interests. And metrics such as executive staff turnover, the size and longevity of funds held by current staff, the size and longevity of funds held by past staff, and the excess of fund returns to compound interest, might shed some light on what insiders really think about fund performance.

Fanciful, perhaps. But how attractive are alternatives such as a regulation induced ‘one size fits all’?

[1] Benmelech R, Kandel E, Veronesi P Stock-Based Compensation and CEO (Dis)Incentives, The Quarterly Journal of Economics, November 2010, 1769-1820, 1770.

[2] Holden R, The Original Management Incentive Schemes, Journal of Economic Perspectives 19 4 (Fall 2005) 135-144, 138.

[3] Even when the scheme was struggling in the Great Depression, the board agreed to defer shortfalls of interest payment and principal for a while but refused to forgive the loans.

18 December 2017