Wednesday, November 11, 2009

Defined contribution plans - a bad idea

There are a number of ways in which people can provide for their retirement. One method which is quite common is the defined benefit contribution plan. Under a defined benefit plan, participants will be entitled to a benefit on retirement which is calculated by reference to factors such as age, number of years in the plan and average salary in the final year(s) of employment. While at first glance defined benefit plans may look like an excellent way of providing for retirement, they are in fact a very bad idea for just about everyone - including the employee.

Here's why:

1. insufficient contributions: in theory, contributions will be made by either the employer or the employee or both over the period of employment. Actuarial calculations will be done to ensure that the total contributions over the life of the plan, together with the return on investing those contributions, will be sufficient to fund the future benefits. With depressing frequency, the contributions are not sufficient. The usual reasons are over estimation of the return on investing the contributions and longer life expectancies. Even a small shortfall in return will make a big difference over the life of the plan. Less frequently, the employer will simply fail to pay - either deliberately or under due to insolvency;

2. who pays for the shortfall: when the value of a plan's assets are insufficient to meet the claims by plan members, one of three things will happen (i) the employer will become insolvent in which case the employees will be left with reduced benefits (ii) the employer will attempt to negotiate lower benefits or (iii) the employer will fund the shortfall - which will usually come as a large and unpleasant surprise to shareholders if (as is just about always the case) the shortfall had not been provided for in the accounts. Another way of looking at it is employees are taking a very significant credit risk on their employer;

3. change of employment: defined benefit plans were designed in the days when people changed jobs less frequently than today. Since defined benefit plans are usually not fully portable, and plan administrators charge significant fees to people who withdraw from a plan), changing jobs or being laid off can have a significant adverse impact on future retirement benefits (remember that the size of the benefit is often linked to years of service). In some countries, regulations have been used to address this problem but with, at best, mixed success;

4. estate planning: if you use an alternative plan (defined contribution or self saving) all the assets in the plan are yours to keep forever (once vested). If you die, you can leave them to your spouse, children, favourite charity etc. With defined benefit plans, there may (or may not) be a benefit to a surviving spouse, but that is all. Once you cease to be eligible for benefits on death your estate has no further claims to either your contributions or your employers. This is fundamental to the way the plans operate (remember the actuarial calculations). If you die a day after retirement, a lot of money gets to benefit strangers. (What happens to your contributions if you die before retirement will vary from plan to plan but you can be assured that you will get less than full benefit);

5. value of benefits: studies have been done as to whether a defined benefit plan produces greater benefits than a defined contribution plan. The results tend to depend on what numbers are put into the calculations and whether a COLA is included. Unions and people who receive defined benefit plans tend to argue that defined benefit plans are better;

6. return risk: someone has to bear the risk of return on the investment of plan assets being lower than what is needed to provide the stated benefits. This should be the employee. Arguments that the employer should bear this risk are logically and morally flawed. As an investor I want to invest in companies that carry on a specific business - not a side business of investing in securities etc. Actually, the employee can avoid the risk by using defined contribution plan assets to purchase an annuity at retirement (and will have flexibility to select an annuity which suits their personal circumstances). From an employee's perspective, taking the investment risk off the table also takes away the possibility of benefiting from above average returns;

7. tax treatment: this is currently a non-issue for me. However, tax treatment in other countries varies;

8. inflation: some plans provide for COLA. Some do not. The value of a non-COLA plan will decline over time (even with low inflation). In contrast, a defined contribution plan can be invested in assets which have at least some prospects for preserving their real value (including annuities).

At the risk of stating the obvious, if one form of plan imposes greater costs on employers than the other, employers will either take the cheaper option or employee fewer people.

As a conclusion, defined benefit plans are generally bad for employers and will be bad for most employees. From the employees' perspective they are effectively obtaining the illusory comfort of a fixed benefit on retirement in exchange for taking on solvency risk, losing the potential for upside returns, losing flexibility with their jobs, increasing the cost of changing jobs or being laid off, losing flexibility on retirement and losing the opportunity to leave something behind for their heirs. If the plan does not include a COLA feature they also face greater exposure to inflation reducing their real post-retirement income. This is a very poor trade off.

The one group of people who may be better off with overly generous defined benefit plans are civil servants whose retirement is backstopped by the the taxpayers. As a tax payer, I find this offensive.

No comments: