Have we locked ourselves into an unnecessarily narrow view of pay for performance? Are we missing opportunities to consider alternate models which are well grounded in labor economics research and literature, but not yet adopted into widespread practice?
That is the assertion of a new Mercer study -- building on the continued work of Haig Nalbantian, Brian Levine and others around this topic -- which calls us to consider three other pay for performance models that rely primarily on base pay to drive desired outcomes.
Tournament Model: A promotion-focused model where pay varies significantly from one career level to the next and employees are motivated by the opportunity to advance. In this model, the best performers earn more via promotions based on relative (rather than absolute) performance evaluation.
Membership Model: Also known as the "efficiency wage model", where pay levels are targeted above the market median and employees must perform to high standards to stay on with the organization. In this model, the desire to keep a highly paid position is what delivers incentives to perform well.
Service Model: Also known as the "bonding model", where employes are motivated by an updward trajectory of pay over time. This model is characterized by pay rates that start below market levels but rise with tenure, offering an attractive "lifetime earnings" potential. In this model, employees are "locked in" over the longer haul, preserving firm-specific knowledge but enforcing performance minimums in order to stay with the organization.
Mercer's study results, which reflect the practices of 570 U.S. and Canadian employers, indicate that 14% of respondents are using a tournament or promotion-focused pay model, 14% a membership model centered on premium pay levels, and only 7% a service or "bonding" model.
Might one of these alternative approaches be a better fit for your organization than the standard merit increase/incentive package? Answering that question demands a careful assessment of fit with your overall talent management model. As always, notes Mercer's Nalbantian, context is everything. These models will succeed or fail based on their alignment with your work environment, your sourcing and staffing practices, your performance management tools and abilities, and your competitive pricing and positioning of pay opportunity, to name just a few key links.
As an example, I would be wary of any interest in pushing one of these approaches based on a desire to let managers "off the hook" for making tough performance decisions. The success of both the membership and the service model hang on the organization's ability to establish and enforce a high baseline level of performance. (I did a three part series on the membership model a few years ago, final post here, that addressed this particular requirement.) The tournament model relies on a relative ranking approach to performance assessment. (See a recent Cafe post on this topic for more perspective on ranking.)
Culture and the time horizon of work being performed might be other important considerations. The competitive spirit underlying the tournament model may make it a poor fit for organizations where collaboration is critical. On the other hand, as Mercer notes, the membership and service models can provide a good fit to circumstances where work is performed and measured on a particularly long time horizon (research and development comes to mind as an example).
An important point highlighted by the Mercer study is that more than one-third (36%) of responding organizations report having multiple pay models in place. Rather than a wholesale replacement of your merit increase/incentive combo model, you might wish to explore one of these three alternatives for a function or unit, perhaps even a job family, where your organization-wide approach is hitting the wall. Maybe the tournament model would be a fit to your engineering group. Perhaps the membership or service model would work well for your scientists. It may be time for a pilot effort and a little experimentation.
More information on the three alternate models can be found in this WorldatWork article on the topic by Nalbantian and Levine.
Readers with experience in (or just opinions on) any of these alternative models - would love to hear your thoughts!
Ann Bares is the Founder and Editor of the Compensation Café, Author of Compensation Force and Managing Partner of Altura Consulting Group LLC, where she provides compensation consulting to a range of client organizations. Ann serves as President Elect of the Twin Cities Compensation Network (the most awesome local reward network on the planet) and is a member of the Advisory Board of the Compensation & Benefits Review, the leading journal for those who design, implement, evaluate and communicate total rewards. She earned her M.B.A. at Northwestern University’s Kellogg School, is a foodie and bookhound in her spare time (now working her way through Siddhartha Mukherjee's "The Emperor of All Maladies"). Follow her on Twitter at @annbares.
Creative Commons image "Three Choices" by Inbarigler
Ann - As I read your post, all three models focus upon base pay as the primary financial motivator of performance. Perhaps a fourth model - I'll call it the Results Model - should be considered. In this model base pay rates are pegged to market competitive rates by function, with adjustments reflecting cost of living increases only. Variable pay offers opportunities to receive additional compensation up to some significant percentage of base pay (50%?) as a result of achieving performance goals as determined by the company. This allows organizations to maintain internal equity while recognizing high levels of achievement.
Posted by: John A Bushfield | 12/19/2013 at 05:48 AM
One big quibble from me, John. Hate to see "cost of living" referenced in these discussions as a valid metric for competitive pay increases. Exactly WHOSE cost of living? Donald Trump's is quite different from mine. And COL measurement criteria vary widely, from various market baskets attributed to wage earners or white collar workers at different income levels in major cities to the Big Mac Index. Would I get a raise if I move from a trailer to a condo? Should pay drop when your kids graduate from college? See my "Of course they want cost of living pay..." article from June 2010 here. Please don't further muddy the already murky waters.
A better term might be "the pay floor," translated as "the minimum competitive market increase experienced by all positions." That usually reflects the typical periodic pay structure changes.
Posted by: E. James (Jim) Brennan | 12/19/2013 at 01:09 PM
Interesting observations, Ann. Thank you. My two cents worth:
A de facto tournament model exists in any situation in which there are indivisible prizes (partnerships, C-level jobs,etc.) and for which there are only endogenous contestants. I'd be hard pressed to see where it could be used in any other situation.
An efficiency wage regime will be workable if you can establish an economic benefit from reduced turnover and increased engagement (or at least reduced propensity to shirk)exceeding the marginal cost of the efficiency wage. (And of course, this begs the question of whether the efficiency wage is in fact the cause of the reduced turnover, etc.) Sam's Club and Costco are running one of the all-time great natural experiments for us in that regard even as we speak. I'd like to think that a future Nobelist in economics is looking closely at this.
I'm dubious about the real-life relevance of the service model in most present-day private sector settings; for it to be workable, employees have to count on their employer's continuing presence in a marketplace in which their skills are valued. The essence of it is that you are under-rewarded (vis-a-vis some market reference point) early in your career, and over-rewarded at a later stage. Given the pace of change (and I see little evidence that it is decreasing), that's a bet that I wouldn't take without receiving some kind of risk premium, which more or less kills the reason for doing it in the first place. Of course, YMMV.
Posted by: Tony Bergmann-Porter | 12/19/2013 at 08:26 PM
Point taken, Jim. I obviously didn't have my thinking hat on securely when I referenced COLA as a basis for pay adjustment. Thanks for correcting me.
Posted by: John A Bushfield | 12/20/2013 at 05:19 AM