There is always much debate about whether being part of a defined benefit (DB) or defined contribution (DC) pension plan is better. People who wish to work for a number of employers and potentially have a number of careers as is often the case these days, and younger workers can often be well served by being part of a DC pension plan. This, of course, assumes that the worker elects to join the DC pension plan. Many are voluntary and participation of workers in DC pension plans is substantially lower than those covered by DB plans (60% versus 90% according to the Confederation of British Industry in a 2006 study).
Putting aside these “softer” considerations, there are some more objective measures that can be used to compare the benefit of membership in a DB or DC pension plan. These include cost, contribution rates and range of return drivers available.
Many studies have looked at the costs paid by DB sponsors versus DC members. The ability to pool assets together has benefited DB plan sponsors as they have larger asset pools from which to draw. U.S. studies suggest that the DC member pays around 1.0% more than the DB plan sponsor. A recent paper by Ambachtsheer and Bauer (2007) indicates that the differential can be as high as 2.0%, significantly reducing the return available to the DC member. The impact of a 1% cost differential for a typical member entering a DC plan at age 30 could be a loss of at least 15% in retirement income.
Many of us in the industry have argued for a greater pooling of assets, more like the Australian system, as a way to increase the asset size and reduce cost. However, there is a reticence in the industry to embrace pooling across plan sponsors. The higher costs paid by DC members become even more noticeable as debt and equity market returns plummet and returns become more anemic.
What about contribution rates? In Watson Wyatt’s 2008 Pension Risk Survey, we found that 62% of Canadian plan sponsors have contribution rates less than 6%. A British study reported similar results—on average, British employers contributed 5.8% towards DC member accounts. This contrasts with British employers contributing 14.2% to DB schemes. A recent Watson Wyatt study finds that U.K. employers have recognized this deficiency and have been increasing contribution rates for DC members—up to 9.8% in 2007, not accounting for voluntary member contributions and matches.
U.S. contribution rates are higher, but still below replacement rates. U.S. plan sponsors contribute 9.8% towards DC member accounts and 19.1% to DB pension plans. Since many DC programs are voluntary and do not require member participation, never mind voluntary contributions, we can see that employees, on balance, are much better off participating in a DB plan, at least using contribution rates as a benchmark.
DC contributions rates are often set without regard for outcome, but rather as a competitive tool. How do current contribution rates impact replacement ratios? Have expectations changed with recent market conditions?
Watson Wyatt tracks expected returns for a number of portfolios, ranging from conservative to aggressive (with the level of equity allocation being the determinant). As seen in the chart below, expected returns (gross of fees and excluding active management) have been falling for the past few years, in part due to declining bond yields:
| Portfolio | January 2002 | January 2006 | April 2008 |
| Conservative | 6.90% | 5.70% | 5.30% |
| Moderate | 7.20% | 6.20% | 5.80% |
| Balanced | 7.50% | 6.60% | 6.10% |
| Growth | 7.70% | 7.00% | 6.50% |
| Aggressive | 8.00% | 7.40% | 6.80% |
If we assume that DB and DC members want to have the same replacement ratio (approximately 50%, not including government benefits), we can follow the impact that declining debt and equity market returns have had on required contribution rates.
Prior to 2002, the required contribution rate for a 30-year-old was approximately 12% owing to strong equity markets. By 2006, this had jumped to 14%-16% to help compensate for lower expected market returns. It will not surprise many to know that with recent market gyrations, the required contribution rate has jumped to 16%-18% in order to effect a similar benefit for DB and DC members.
It does not take a mathematician to see that current DC contribution rates in Canada, the U.S. and U.K. are far short of these amounts. (These amounts are much closer to the contribution rates for DB plans, though).
The increasing contribution rate also highlights another deficiency of the DC plan—a heavy reliance on the equity risk premium. Yet another pet peeve of mine is the fact that DC members do not have the same investment options available to them (as DB plan sponsors), even within a portfolio context. While it is not reasonable to expect that most DC members have sufficient expertise to invest in hedge funds, private equity and the like, adding more diverse return drivers (such as credit, skill, illiquidity, etc.) from within a portfolio context could lessen the members’ reliance on equity markets alone. This could be delivered through the Australian model mentioned earlier in the article and touted by Keith Ambachtsheer. As a former contributor to an Australian DC program, I can personally attest to many of its virtues.
Has this article furthered the debate as to whether membership in a DB or DC plan is better? I will leave you to be the judge of that. It does highlight some of the disadvantages that DC members are exposed to—higher cost, lower expected returns and a heavy reliance on the equity risk premium.
We as an investment community need to figure out a better way to deliver the DC benefit. Simply saying DB plans are better is too simplistic. However, it is clear we can significantly improve the design and delivery of DC plans—we includes the investment consultants, investment managers, plan sponsors, administrators and record keepers. We need to figure this out soon as well. Globally, DC plan assets are forecast to eclipse DB plan assets by 2014.