Thursday, September 07, 2006

DC Plans Underperform DB Plans by 1%

A recent study by the Center for Retirement Research concludes that DB plans have outperformed DC plans by at least 1% since 1988 and that lower DC plan returns don’t fully reflect the investment allocation problems at the individual participant level that may further negatively impact their preparation for retirement.

Thought the comparative asset allocation equations and return calculations were fairly imprecise and in some cases inconsistent with other sources(i.e. DC equity % allocations vary from 50% - 67% over the last 10 years vs. EBRI data suggesting DC equity has “varied narrowly (62% to 77%) over the past 10 years), we are not surprised by the overall return conclusion and, in fact, would expect the DC underperformance to be even greater. In addition to the fee structures there are probably several other fundamental reasons why DC plans should, on average, return less than DB plans:

  • Underutilization of Risky Assets – DC plans as a matter of liability control and to avoid participant confusion do not generally offer hybrid or high risk asset classes such as; emerging markets equity, real estate (REITS), or high yield bonds. In a disciplined asset allocation exercise characteristic of DB plan management these asset classes are often included to maximize returns/risk adjusted returns. Over the study’s 16 year horizon, the S&P 500 outperformed all of these other asset classes – narrowing the expected impact of this asset allocation differential.


  • Dependence on money market & stable value Products – DC plan fixed income allocations are predominantly in money market or stable value products vs. DB plans which use market value bond portfolios. Money market funds underperformed intermediate bonds by almost 4% annually from 1998-2004. Stable value products should be expected to under-perform intermediate bonds by some amount due to the “insurance” cost.


  • Indexing – Index fund management became firmly established in DB pension plans in the 90’s. The allocation of DC assets to index funds we suspect is lower than that of DB assets. This is likely reflected in both lower costs and the possible avoidance of negative alpha which is so prevalent in active management.


We also do not find that the author’s data detailing participants lack of diversification lead us to the conclusion that “most participants face the risk of ending up with inadequate retirement income”. In fact, the lack of diversification in these plans may reflect very rational participant decision making for at least two reasons.

First, the universe of plans in the study was constrained to those which included had complimentary DB and DC plans. Under these circumstances, a DC participant who also anticipates a DB plan benefit should consider the DB benefit a fixed income asset and therefore overweight their DC allocation to equity. For long tenured employees with a good accrued DB benefit, 100% equity allocation in a DC plan would not be unreasonable.

Secondly, while we can observe the asset allcoation for the entire pool of assets dedicated to retirement liabilities in a DB plan, DC asset allocations often reflect only a portion of participant’s assets dedicated to retirement. Tax differences may skew how retirement assets are held. For example, it might be more tax efficient for higher income employees to buy and hold their equity asset in taxable accounts while holding their fixed income or stable value assets in tax deferred accounts to avoid higher current tax impacts.

0 Comments:

Post a Comment

Links to this post:

Create a Link

<< Home