Why We Need A Living Wage Plus Parity

(Why HCECP's Parity Proposal Isn't Enough)

The HCECP report released on December 19 confirmed what our community has known for years: Harvard's service-sector workers are paid far too little. At a time of astounding institutional wealth, Harvard has cut real wages for guards, janitors, and dining hall workers, leaving employees unable to afford basic necessities without working multiple jobs and sacrificing their health, family commitments, and responsibilities outside work. The harm that Harvard's low wages do to workers is immense and unjust; it is an assault on their dignity and humanity.

The HCECP report revealed not only the spread of poverty-level wages at Harvard, but the dynamics of power and authority at the university which have pushed so many workers into poverty. The report documented the wage cuts that Harvard workers have suffered in contract negotiations since 1994, testifying to the gross power imbalance that workers face when bargaining with an $18 billion corporation. And the report confirms that Harvard has used its disproportionate power to evade responsibilities to which it ostensibly commited itself, as when it failed to implement the report of the Mills Committee. The HCECP findings demonstrate that Harvard's administration has systematically abused its power in order to immiserate workers, and this fact should prompt us, when evaluating any policy recommendation, to consider not only its theoretically predicted effects but the ways in which it will be affected by the power imbalance that we know to exist at Harvard.

There is agreement in our community, and there was unanimous agreement within the HCECP, that Harvard must raise wages throughout the service sector and ensure that they never again erode to poverty levels as they did over the last ten years. Yet the committee's recommendation of a one-time wage boost and subsequent parity -- pegging wages for outsourced and non-union workers to the wages of directly-hired, unionized workers -- clearly fails to meet this goal. In fact, no policy lacking an annually-adjusted living wage floor will meet this goal. Below, we look at a number of policy configurations that might aim to raise wages and provide long-term protection against erosion. We argue that the only effective configuration, absent a ban on outsourcing, is a combination of parity and an annually-adjusted living wage policy.

Our argument embraces part of the HCECP report -- we maintain that parity is an essential component of a fair labor policy -- but we explicitly reject the report's framing. The HCECP report portrays an annually-adjusted living wage policy and a parity policy as conflicting proposals between which Harvard must choose. We believe that parity complements a living wage policy naturally and effectively, and we join eight of the HCECP's 19 members who wrote separate statements in the report to say that this dual policy should have been the committee's recommendation, and should be the university's policy.

I. There is no effective mechanism to maintain wages above poverty levels in the future other than an annually-adjusted living wage policy to accompany parity.

Eight of the 19 HCECP members agreed with the expressed community consensus that an annually-adjusted living wage policy was the only mechanism that can reliably maintain wages at levels that will meet the local cost of living. Significantly, these eight members included the majority of the workers and students on the committee -- the only categories of HCECP's membership not appointed by the administration -- as well as every person of color on the committee.

The refusal of the other committee members to recommend a permanent living wage policy, juxtaposed with their stated goal of guaranteeing that wages never again fall to poverty levels, suggests that there are other equally good ways to raise wages and maintain them indefinitely into the future. Below, we consider some possible mechanisms and argue that, even if implemented with parity requirements, none of them could provide the same wage protection as an annually-adjusted living wage policy. In fact, they all invite future wage erosion.

  1. Consider HCECP's recommendation: parity with an across-the-board, one-time wage increase.
  2. A one-time wage increase can be counted on to erode over time. For instance, the guards' union has been decimated by Harvard's deliberate union-busting, and would at this point be hard pressed to defend a wage increase in its next negotiation, let alone improve upon it. More generally, the bargaining power of even the strongest union can be counted on to increase and decrease over time, and when they are weaker, all unions will lose ground on wages. A one-time raise without future adjustment frankly accepts that in the long run, workers' ability to support themselves and their families will be left to changing power relations that have no right to determine the conditions of people's lives.

    Some might say that after implementing a one-time wage increase, the university might treat the increase as a precedent and not demand concessions in future negotiations. We find this suggestion incredible given Harvard's history of bad-faith negotiation. In fact, it is fair to predict that the only reason that Harvard would prefer a one-time increase to an annually-adjusted living wage is because it hopes to erode the one-time increase over time.

  3. Consider: Parity with a floor set at the level of the university's lowest collectively bargained wage.
  4. There is no guarantee that the lowest collectively bargained wage will be fair or in line with the local cost of living. Right now, the lowest collectively bargained wage at the university is roughly $8.50 per hour -- the starting wage for directly-hired museum guards. And Harvard's decimation of the guards' union strongly suggests that the university will be able to keep their wages low unless an added mechanism is adopted to raise and protect wages.

    Setting a floor at the lowest collectively bargained wage in fact creates perverse incentives for the university to use coercive negotiating tactics to keep starting wages extremely low. Harvard's ability to do so should not be underestimated: when bargaining with weaker unions, the university already does bargain low wages. And even if all Harvard's unions were very strong, Harvard would likely push for wage structures in which new hires worked for a very short period of time at a very low rate, so that the university's wage floor would remain low; workers in the union might then quickly receive raises to satisfy them and the union while the university-wide wage standard would remain very low. Clearly this would be destructive and coercive, yet it could be quite difficult for unions to resist powerfully, since very few of their members would ever be making the abysmally low starting wage. Meanwhile, the low starting wage would be the floor for all outsourced and casual workers.

  5. Consider: Parity with a prevailing wage agreement, or an agreement that Harvard's wages will be set at, for instance, the 90th percentile of the distribution of wages in the market.

A market-based wage standard should be unacceptable to us for a number of reasons.

    1. HCECP was established because of our community's assertion that there are standards Harvard should meet because they are just, not simply because other employers meet them or because Harvard is forced to do so by larger trends. In other words, it was our community's expectation, and HCECP's task, to produce standards for Harvard that are not market-based, but principle-based.
    2. If we are to consider policy options based on principle, we must recognize that the fundamental problem with Harvard's low wages is not that they compare badly to wages at other institutions, but that they immiserate workers and fail to provide for essential human needs. Policy recommendations should reflect the principle that an institution with the resources to do so should pay workers enough to meet basic costs of living, not the principle that Harvard should do what other employers do, or even the principle that Harvard should do somewhat better than what most other employers do.

    3. There is no guarantee that prevailing or market-based wages will be adequate. They could still leave workers in poverty. This is especially true in the service sector, where wages are notoriously low. Prevailing wage agreements are effective in the building trades, for instance, because that sector is densely unionized and unions have established high standards as the industry norm. But the service sector is not densely unionized, and so a prevailing or market-based wage standard would not afford the same protection.
    4. A prevailing or market-based wage standard is conceptually ambiguous, and can lead to unfair managerial manipulation. For instance, there is no simple way to choose an appropriate market. Should Harvard's janitorial wages be set to janitorial wages in the Boston area, to janitorial wages at Boston area universities, or to janitorial wages at institutions of similar size and wealth? Of course, the administration can be expected to choose the market definition that would yield the lowest wage standard for itself.
    5. A prevailing or market-based wage standard would prove almost impossible to monitor and enforce. Ultimately, labor standards must be able to be monitored by the community; the standards and implementation must be as transparent as possible and violations need to be simple to identify. Yet it would hardly be easy for a member of our community to know whether the wage of a worker she was speaking with was indeed at the 90th percentile of the distribution of wages in the relevant market.

II. An annually-adjusted living wage policy is an effective and sound complement to parity.

Parity with an annually-adjusted living wage policy will raise wages and protect against future erosion. This much should be clear, and this alone puts a parity plus a living wage policy head and shoulders above the options listed above, which do not meet this goal. The only question, then, is whether a living wage policy would have so many negative ramifications that they would outweigh its clear benefits. Consider, then, the following possible concerns about a living wage policy -- concerns which were presented in quite a biased way in the HCECP report:

Concern 1: "A living wage standard will become a ceiling in union negotiations."

This concern might refer to two different circumstances. First, in a workplace without parity, competition between directly-hired workers and outsourced workers could conceivably make a wage floor into a ceiling. However, a parity agreement would eliminate this problem, as all HCECP members agreed. This scenario demonstrates the complementary nature of parity and a living wage policy.

In the second scenario, Harvard might implement both parity and a living wage policy. Here, the idea that the living wage standard would become a ceiling seems highly counterintuitive, runs against most theories of bargaining, and is a concern that has never been voiced by any part of the labor movement. Campus unions, as well as their international affiliates, have endorsed a living wage policy because they say it will help them negotiate higher wages -- they will not have to devote as much of their bargaining power just to raising their lowest-paid members out of poverty, and can fight for more substantial gains.

The most recent negotiations for Harvard's dining hall workers provide an example of how a living wage policy could have helped a union win higher wages. The dining hall workers' union, HERE Local 26, decided that its first priority was to raise the incredibly low wages of "cash operations" workers -- the lowest tier of food service workers on campus, created by Harvard in the 1990s. The union was able to raise their wages, but even when bargaining from a position of strength, it took virtually all of the union's bargaining power to do so. Local 26 was only able to get raises for higher-paid "board operations" workers of 20 cents every 6 months because it had to put all of its energy into fighting for what was barely a living wage for its lowest-paid workers.

It is certainly true that Harvard will view any wage floor as a standard to try to push people down to. But that is already the case: over the last ten years, Harvard has pushed workers down to the lowest standards possible -- whether it's Sodexho food service workers who are now hovering just above the Massachusetts minimum wage, or directly-hired janitors forced down to the standards set by ruthless contractors. Instituting a high living wage standard and making sure it is annually adjusted will at least raise the standard down to which Harvard can try to grind workers, and it will give workers and their unions a decent place to bargain up from.

Concern 2: "A living wage policy will shift worker skills and demographics."

Some HCECP members claimed that higher wages would attract a workforce that was whiter and more highly-educated than the current one; they warned, therefore, that a living wage policy would change Harvard's pool of workers, hurting the people it was designed to help.

We have seen no solid evidence showing the inevitability of skill or demographic shifting. Research presented to the HCECP by the Harvard Workers' Center indicated that upper bound estimates on such shifts would be very small. Comparative demographic information from other area universities which pay substantially higher wages substantiate this claim. Moreover, it is clear that the university could implement protections against shifting -- from worker retention policies to affirmative action measures.

More importantly, this complaint is not an objection to a living wage policy: it is an objection to higher wages. If HCECP and Harvard intend to raise wages in any way at all -- which at least HCECP claimed it wanted to do -- this criticism could be leveled at whatever they suggest. In that case, issues of skill and demographic shifts should be discussed and addressed, but they should not be used as an argument to dismiss a living wage policy.

Concern 3: "It is very hard to pick a number."

The HCECP report asserts that some members could not agree to a living wage policy because they found it too complicated to select a number. Any wage, they noted, would yield different results for families of different composition, and therefore could not be a perfectly-calculated standard.

This is certainly true, but it is no reason not to pick a number. HCECP gathered a range of living wage estimates produced by the Economic Policy Institute and the Women's Educational and Industrial Union, and heard testimony from economist Robert Pollin on ways to set a living wage standard. The wage estimates from EPI and the WEIU addressed the needs of families of differing compositions in our area, and fell mainly in the range $11 to $22 per hour. Harvard not only could, but should be expected to select a living wage standard from this range. Its judgment in choosing a particular number might be guided by two considerations. First, what is Harvard's ability to pay? Clearly Harvard is wealthy enough that it could select quite a generous number. Second, what might be the larger social effects of Harvard choosing a given number? This consideration might restrict Harvard from choosing a very low number -- which would have far-reaching effects in setting a low standard for other employers and universities -- or, in the opinion of some, from choosing a number at the top of the range, which might be criticized as threatening negative economic impacts. Such reasoning could easily support the adoption of a figure of $12 per hour or more. Such a standard might be further supported by comparison with equal and higher wages paid in the past in Harvard's service sector, and paid today for service-sector jobs at other area colleges and universities.

It is true that this reasoning does not guarantee wages determined "scientifically" -- if any wages can be said to be "scientifically" determined -- to precisely meet the needs of every worker and family. Twelve dollars per hour means different things for different families, and for people whose lives vary in many other ways. Yet for HCECP members, and for Harvard, to refuse to choose this or any other number on the basis of such variation is to accept that Harvard will continue to be restricted only by another incomparably worse number: the state minimum wage of $6.75 per hour. The experience of Harvard workers, some of whom make very little more than that, shows clearly that this is a number which is inadequate for any worker and family. And derived as it is from political negotiation, it is a number far more arbitrary than any chosen from cost-of-living studies.

Whether Harvard adopts a living wage policy or not, then, it will be choosing a number. We remain convinced that one selected from a range of cost-of-living figures is far more thoughtfully chosen, and far better for workers, than the default floor of the Massachusetts minimum wage.