In asset allocation, union plans indeed have been more conservatively invested than corporate defined contribution plans (in terms of the percentage of assets allocated to equities) and historically had slightly lower investment returns.
However, data compiled by the Employee Benefits Research Institute, Washington, indicates asset allocation trends are converging among single-employer defined contribution (whose allocation to equities in 1990 was 45.9%) and defined benefit plans (41.7% equities) and union plans in general (38.7% equities).
In terms of investment performance, multiemployer plans stack up well relative to corporate plans, with five-year average returns only one percentage point behind corporate defined contribution plans and 0.1% ahead of corporate defined benefit plans.
Shorter-term data indicate an even wider spread, with union funds earning slightly more than corporate plans for the third quarter 1993, 3.9% for union plans and 3.7% for corporate defined benefit plans.
But union members in these plans cannot take all of the credit for the good news - just as they shouldn't take all of the blame for any bad.
Therefore, it is important to note at this point that there exists no situation in the private-sector pension system where union leaders sit as sole or majority trustees over pension assets. In multiemployer plans - which is the only setting where union members act as trustees as a matter of law - union members make up only half of all trustees; the employer sponsors make up the other half. So there are practical limitations to "labor's paternalism."
But asset allocation and investment returns don't tell the true story of union plans. The question is not whether they are earning as much as corporate plans, but whether their returns are appropriate for their ability to tolerate risk. To answer this, one must understand how union plans are created, how contributions are determined and how this affects benefit payments.
For example, unions bargain for many employers to contribute to the fund for every hour and day members work. This money is held in trust and can only be used for participants. Employers cannot adjust how much they contribute to the plan unless they negotiate with the union.
On the other hand, sponsors of corporate plans can alter contributions as long as their funds fall between the wide funding guidelines set forth by the Employee Retirement Income Security Act of 1974.
All of this means investment shortfalls in union plans cannot automatically be made up by obtaining more contributions from employers, as in corporate plans, but must be obtained, instead, through the collective bargaining process. This structural difference is not academic - it has meant real differences in retirement income security.
As a result, the risk-averse nature of union plans is appropriate and in the 1980s union plans did earn slightly less than their corporate counterparts.
However, the Department of Labor's John Turner and Stuart Dorsey showed in a 1990 academic publication that union plans were safer so that their risk-adjusted rates of return actually were better than corporate plans.
In addition to their high and stable rates of return, multiemployer union plans distinguish themselves by having labor and management as equal partners on the trustee board.
Second, multiemployer trusts have been in the forefront of economically target investment for more than 25 years.
In fact, the DOL and Department of Housing and Urban Development have been working with the AFL-CIO Housing and Building Investment Trust to put money in low-income community houses in rebuilt and rebuilding neighborhoods. The rate of return has been a whopping 11%. The Bricklayers union earned an above-market rate in the late 1980s by investing in low-and moderate-income housing in Boston. The Sheetmetal, Boilermaker, Carpenters, International Brotherhood of Electrical Workers, and indeed almost all union plans, have some ETIs.
The Turner and Stuart study concluded the multiemployer commitment to getting a good return and some other economic and social feedback benefits does well for participants.
But the great divide between union and corporate plans is what they deliver to participants - a much more direct and important gauge of how different these plans are for workers.
Pension funds earned huge gains in the go-go financial markets of the 1980s. Between 1983 and 1990, pension funds more than doubled in total assets (going from $854 billion at the end of 1983 to $1.723 trillion at the end of 1990). What corporate and union funds did with that windfall is vital.
In a University of Notre Dame study, Teresa Ghilarducci finds multiemployer plans accrued more than $400 per participant in 1981, while corporate funds accrued more than $700 per working participant. By 1991, union plans had increased pensions at a higher rate than corporate plans and surpassed them in generosity. In 1991, the accrued amount for working participants in union plans was $1,085, while for corporations it was $938.
In the 1970s, pension funds regularly gave cost-of-living adjustments to retirees. By 1980, corporate plans (outside of collectively bargained plans like the United Auto Workers' negotiated plans with the Big Three automakers) practically stopped all COLAs for retirees.
In contrast, giving COLAs to retirees is a standard practice for union funds. The Sheet Metal Workers even fund them in a separate COLA trust. We do not know of any significant corporate plan that guarantees COLAs.
Corporations might have earned high rates of return in their pension funds, but it didn't go to pension participants.
Union pension funds follow the traditional democratic principles as well as the modern principles of worker participation. No multiemployer plan has been raided and terminated like thousands of corporate plans in the 1980s. Some multiemployer plans have surplus assets, while many corporate plans are severely underfunded.
And, lastly, it should be apparent that no pension plan, no matter how well invested, benefits an individual if that individual is not participating in the plan. By this measure, unions win hands down.
According to EBRI tabulations of the April 1993 Current Population Survey, the participation rate for union covered workers is 78.7%, while for workers not covered by a union participate it is only 40.5%.
As for Pensions & Investments trying to teach the lessons of the Pullman strike, let us say this. Just two of the lessons the long history of trade unionism teaches are: Give workers the dignity of a democratic voice in matters that affect their lives and you'll have a responsible worker and, promise only what you can deliver and deliver what you promise.
If this is paternalism let us have it. Perhaps P&I should consider recommending all pension funds be in protected jointly managed trusts, where investment gains (motivated by risk aversion or not) go to pension participants.
Teresa Ghilarducci and Richard Grant are, respectively, assistant director and senior policy analyst in the department of employee benefits at the AFL-CIO in Washington.
