Time for a Reality Check on Salary Ranges?

With 2010 salary increases expected to come in at an average 2.5% and experts questioning whether they will return to previous levels anytime soon, it's time to rethink salary ranges.

Base salary ranges are administrative tools whose purpose is to guide pay decisions in line with an organization's compensation philosophy. When well designed and used, they impose order on base salaries.   They help ensure that employees are consistently, competitively and fairly paid.

With the exception of broadbanding and other alternative approaches, salary range spreads (the spread from range minimum to range maximum) have followed pretty consistent protocols over the past few decades.  From George Milkovich and Jerry Newman's widely regarded textbook Compensation:

Top-level management positions commonly have range spreads of 60 to 120 percent; entry to midlevel professional and managerial positions, between 35 to 60 percent; office and production work, 10 to 25 percent.  The underlying logic is that wider range spreads in the managerial jobs are designed to reflect the greater opportunity for individual discretion and performance variations in the work.

In my experience and consistent with the above, the most common salary range spread is 50%, with minimum and maximum set at 80% and 120% of midpoint respectively.

Range architecture like this made sense when it first caught hold, back in the 70's and 80's, a time when average salary increases were 8-10%.  With salaries moving at that speed, employees could expect to progress through their salary ranges at a pace which matched their personal skill and performance development.

The times, however, have changed ... and so has the speed of base pay growth.  And yet many - even most - organizations are managing salaries with ranges similar to those described by Milkovich and Newman.  Which means there is (and, frankly, has been) a disconnect between the reality of base salaries today and the tools we are using to manage them.

Consider this:  If the delicate dance of salary structure adjustments versus salary increases continues its pre-recession pattern whereby ranges are typically adjusted annually by an amount equal to 60-70% of the average budgeted salary increase, then projected 2010 increases of 2.5% mean ranges will be adjusted by an average of 1.75%.  A scenario in which the average employee makes about .75% headway in their respective salary range.  A scenario, when coupled with a salary range spread of 50%, in which it would take the average employee about 25 years to progress from the range minimum to midpoint (typically the "market rate" for their position).

Sound like a winner to you?  Me neither.

The answer?  If salaries stay on the low growth path (and inflation doesn't enter the picture to alter the paradigm in any substantial way), I think we'll be forced to consider narrower ranges - ranges which support a more reasonable salary progression relative to the market or "target" value for the job.  It may also, however, be time to begin fundamentally rethinking the kinds of tools we use to manage base pay.

Your thoughts?

Comments

[Vita Taylor]

Here's a new (or not so new thought)...
How about ditching the ranges and moving to market indexes ((individual pay/market rate)*100). They may provide a clearer picture of the salary market to management. I realize it is more work, but maybe now is a good time to reevaluate the system as a whole.

[Harve Presnell]

Don't many employees receive promotional increases of 6-8%, especially early in their careers, that move them through the range faster than you stated?

[E. James (Jim) Brennan]

Correct, but not at all new. I'll send you my article of exactly 30 years ago: The problem with salary ranges (and a new solution), Personnel Journal, March 1980.

Agree with Vita, too. Fixed ranges reflect followship rather than leadership, showing that habit trumps sense.

Harve's right that most pay growth comes from promotions rather than same-job increases; but that places them in a brand NEW range where they typically fall behind due to the compounding effects of basing merit bumps on incumbent salary rather than the central job value. More senior peers in the newly-promoted job have higher bases so their pay increases are bigger in absolute terms, etc. It's an old issue, too.

[Nancy A, Philadelphia PA]

Re-thinking tools is always a good idea. I have always been and continue to be a fan of "market pricing", often using 25-75%iles as reasonable guides to providing competitive salaries. Some might also use 10-90%iles as a way to manage salaries within the market, excluding outliers.

[Ann Bares]

A busy week has delayed my response to the wonderful comment stream going here - but I have been watching and enjoying the discussion.

Vita:
Good point. I have several clients who follow this approach, and you're right - it provides the anchoring advantage of traditional ranges, yet leaves you flexibility in how tightly you control around that anchor. Great food for thought.

Harve:
As Jim points out, a promotion (presumably to a higher level job and set of responsibilities) does in fact put them in a brand new range where - in many cases - they start the climb through all over again.

Jim:
If we stay in this field long enough, it is interesting to see old ideas become new again (and vice versa) as the economy dishes out the newest set of problems and everyone scrambles to react. Or maybe only you and I think so ....

Nancy:

Agreed - always need to keep re-looking, re-thinking and revising our tools!

Thanks, all, for the thoughts here!

[Margaret Dyekman]

Very thoughful comments here. I will add that slotting jobs in grades is useful for internal equity clarity and forces a look at job responsibilities and work design. Sometimes organizations need that exercise to help bring order out of chaos across departments and functions. With grades typically come ranges, and I always recommend an analysis on 25th/50th/75th be done on benchmark jobs before determining range spreads. A 50% range spread across the board raises a red flag for me.

[Ann Bares]

(The following posted on behalf of Joe Brown, in response to the hard time Typepad decided to give him. Thanks for perservering, Joe, in your attempt to share your thoughts here.)

As you know, Ann, I always find your pieces compelling, even when they bring out the compensation geek in me. You've raised some good questions about the use of salary ranges in light of the prevailing economic conditions over the past couple of decades, but I am of the opinion that, at least in the sector in which I consult, they still have utility. I do agree, however, that there are some tweaks in their design and management which could be of value.
I cut my teeth as a consultant with the Hay Group, working across industry sectors, and salary ranges were a near universal component of the compensation installations that we worked on. (Although we did at dimes work near the other ends of the salary management continuum -- maturity curves and broadbanding.)
For the past 12 years, I have worked exclusively with nonprofit organizations. This sector, perhaps because it tends to be dominated by smaller, less well-resourced organizations, tends to lag the for-profit sector in terms of human resources practices, but its needs can be just as challenging -- if not more so, given that they typically have more diversity in terms of constituencies. In my work in this sector, I have found that salary ranges are still quite important as the basis of rational compensation approaches. While this may come as a surprise to those working in the for-profit world, salary ranges and associated practices are often a step up from nothing for many organizations.
There area a couple of things that I have done in my work with nonprofit organizations to keep salary ranges useful and effective.
One, unlike the practice you describe of moving ranges each year by 60% to 70% of budget salary increases, is to slow range growth, encouraging most organizations to review and potentially move their ranges every 2 to 3 years. This not only simplifies things and cuts down on the time and resources involved with market reviews, but also means that employees experience more progress through the ranges.
As you pointed out, it is definitely harder to differentiate rewards by performance in the 4% world we've lived in for the past couple of decades (or now, perhaps, the 0% to 3% world). One response to this has been to use fewer categories of position-in-range (or compa-ratio, though I've always hated that term). The organizations I work with typically use three categories, which allows for slightly greater differentiation and the ability to move those low in the range up a bit faster.
Your analysis of the average employee taking 25 years to progress from minimum to midpoint in the typical range ignores one point that I feel is very important: Over any significant period of time, most employees will or should be promoted to higher levels, or see their jobs grow to higher grades. For those who do not -- presumably because their performance is subpar -- it is entirely appropriate that they remain low in the salary range. I strongly reject the notion that all employees should, as a matter of course, progress towards midpoint, regardless of their performance.
You mentioned the possibility that we need to start using narrower ranges, but I don't see a strong argument for them in your post. As I see it, they would not change the market orientation of salary ranges, and while they would actually increase the consistency aspect of internal equity, they would in fact diminish the ability to differentiate pay on the basis of performance.
I would introduce another possible alternative into the conversation: asymmetrical ranges. To my knowledge, there is no rule that salary ranges have to be symmetrical around the midpoint, although it is clearly our most common practice. I recall at least one time in the past working with an organization to construct asymmetrical ranges. For example, a salary range of 90% to 120% would mean hiring people closer to the midpoint, and having average performers progress toward it faster, but would also provide more room to reward high performers.
Again, thanks for a compelling post, Ann. It actually inspired me to write a two-part series of posts on salary ranges for nonprofit organizations. The first part was published today on the Mission Connected blog, http://blog.execsearches.com/2010/04/06/salary-ranges-part-1-why-ranges/, and the second part will appear next Tuesday. I'd definitely be interested in your feedback on the posts or my thoughts above

[Ann Bares]

Margaret:

Thanks for the comment. Sounds like you're advocating tailoring range spreads of different grades to the specific set of pay data. Interesting position. Would be interesting to hear your logic (why a consistent range spread is a "red flag" to you) - as I would lean strongly in the opposite direction for a number of reasons. Hmmm, maybe a future post topic...

Joe:

Wow - lots of commentary. While you and I share our strong regard for the median, we may be at odds in other areas. First all, I certainly don't mean to advocate for getting rid of salary ranges - I am, in fact, a big fan of salary ranges. I only mean to suggest that, given the circumstances in which we find ourselves, it might be time to think about doing them differently. A few reactions, as you request:

Slowing range movement down by only adjusting them every three years? Well ... yes, that would certainly allow employees to progress more quickly, but they would then be progressing through salary ranges that were no longer in sync with the market. Not sure we've solved a problem with that approach.

I also don't mean to advocate for all employees advancing to midpoint, regardless of their performance. The sample scenario I outlined simply dealt with an average employee who is earning a "budget" or "market" level salary increase. Presumably we wouldn't be awarding that level increase to non-performer, although stranger things have happened....

Narrower ranges don't change the market orientation of the ranges, they merely tighten control around the market value. And yes, a narrower range, by tightening control in this way, provides less flexibility for rewarding (or not) performance - so there is an element of trade-off to be addressed. (In fact, salary range design is a process full of trade-offs...)

Asymmetrical ranges are a solution worth exploring, as I did in a post a few years ago...
http://compforce.typepad.com/compensation_force/2007/10/align-your-sala.html

... although here again, I think we need to tread cautiously and be aware of the circumstances we are reacting to in our design - so that we can revisit our design if and when they change.

Anyway - just a few quick thoughts back. Appreciate the thoughtful comment and the link back to the series on your blog!

[Joe Brown]

Hi Ann. Thanks for posting my comment and for your reply. I don't think we're actually so much at odds. The only of my points you seem to take issue with is slowing range movement down to every 2 to 3 years instead of annually. I understand your point about this, and the approach may mean being less "in sync" with the market, but I personally don't view any market data as being so precisely reflective of reality that having a midpoint that is 1% to 3% off of it at any given point in time puts one significantly "out of sync" with it, particularly when dealing with a +/- 20% range. At today's growth rates, we are talking about the difference between moving ranges 1% to 3% a year versus moving them 2% to 6% every two years or 4% to 10% every three years. Also, as noted, most of my work is with small to mid-size nonprofit organizations. For them, the administrative savings associated with going through the range adjustment process less often can be considerable and perhaps worth any temporary and theoretical "slippage".

Seems we are very much on the same page with regard to performance and range movement, as well as our understanding of the dynamics of narrower ranges and the trade-offs of salary range construction.

Thanks very much for sharing your earlier post on asymmetrical salary ranges. Very thoughtful look at situations where they may be appropriate. Definitely think I will move it closer to the top of my bag of tools..

Just FYI, the second of my salary range posts on the Mission Connected blog was published today, earlier than I expected. It is here: http://blog.execsearches.com/2010/04/08/salary-ranges-part-2-anatomy-of-a-range/

Thanks again for the reply, Ann. Love talking comp with you.

[Ann Bares]

Thanks, Joe, for the follow-on thoughts and the link to the second post. I also work with a number of nonprofits, including some smaller ones - so I appreciate your position on their particular circumstances and needs. Great convo!