Count the Zeros or Discount the Zeros — That Is the Question
July 28, 2009 — A few weeks ago, WorldatWork published high level data from our 2009-2010 Salary Budget Survey. A lot has been made about the 2.2% salary budget projected for 2009, and how it is a drop of 1.6 percentage points from the 3.8% of 2008. I have read many articles quoting our survey and others discussing the disturbing trend of lower increases and projecting those increases into the future — the "new normal."
Now, I am struggling with this data. The SBS is accurately reporting the average increase at 2.2% — however, that number includes all organizations that are reporting 0% increases this year. In previous years participants who projected no increase (0%) numbered only 2-3% of total respondents, but this year that number jumped significantly (30-40%+ based on employee category). By removing the 0% respondents, the average increase (of those giving an increase) sits around 3.1%. Now, 3.1% is still less then 3.8% of yesteryear — but it is still significantly better than 2.2%.
I understand the compensation theory of including 0% and looking comprehensively at how the market is moving — but — if you are one of the 2/3 of organizations budgeting for increases this year, wouldn't you be ignoring your due diligence if you did not look at the increase without 0%? I am not recommending ignoring it completely, but, it is important to look at the following to get a better view of market movement:
1/3 of organizations plan 0% salary budget in 2009
2/3 of organizations plan 3.1% salary budget in 2009
100% of organizations look at a 2.2% salary budget in 2009.
Jim, am I off base?
JC, I think you are right on target. I think the larger point you are making is that you can not use SBS data, or macro market data in general, without understanding the underlying issues and trends. You also have to make sure your data reflects your peer group and your labor market. Macro data by itself is headline news, but it doesn't add much value beyond that when it comes making value-added, data-driven decisions about something as important as salary budget decisions.
We have summarized before the data hierarchy that should go in to a merit budget recommendation. But it would be worthwhile to review this again. We recommend beginning with the macro labor market information as a foundation; this would include things like SBS numbers, inflation and employment cost information, unemployment and general labor market availability. Then we would look at internal business data; things like turnover, time-to-fill hiring trends, hot skill concerns, strength of business performance, and similar data related to an organization's particular circumstances; and finally, and I would say most importantly, we would look at market-pricing for particular occupational groups. Here is where the important work of benchmarking, survey analysis, and salary structure and pay policy lines evaluation comes in to play.
Of all the work that is done in developing a salary budget, this final step of determining what the actual market is for specific positions and what the pay policy will be in order to attract and retain talent is the most challenging and the most critical step. The other data can be obtained off-the-shelf more or less. But you can't do market-pricing without doing your homework. This will make the issue of counting or discounting the zeroes a moot point because you will know what is actually happening in the market-place for the talent your organization relies on.
Rodney B Hulsey, CCP, PHR Compensation Consultant Internal Member Since: 2/10/1997 Comments: 12
Thank you Mr. Dobyns for bringing the topic back to professionalism. I was as dismayed as you were that such 'cherry picking' was suggested as an option. The role of compensation professionals is to report the data. If the CEO chooses to remove the zeros, that's the CEOs prerogative. But the compensation professional should never make that choice without making it very clear that they are not reporting the data as reported - they are reporting as they chose to report it.
Doug Sayed, CCP, SPHR Principal Member Since: 2/13/1991 Comments: 10
Good discussion. The responsible thing to do in this environment is to report it both ways (with and without the 0% respondents). This way we can make a more reasoned decision. Also, knowing the % who say 0% would be worth knowing.
I agree that in more normal or better times, this isn't a big issue, but these are not normal times.
Paul Weatherhead Program Manager Member Since: 5/1/2000 Comments: 467
Maybe its time to start reporting the numbers as a range of responses instead of a single average? For example, maybe the best way to report the numbers is like this: (using made up numbers for illustration only)
Christopher Dobyns, CCP Deputy Manager, Office of HR Strategies Member Since: 1/1/1998 Comments: 11
Although I rarely offer comments on the blogging topics, I guess the angle on this subject caught my attention on a couple of different levels.
Exclusive of the good things Jim highlighted, I guess I’m more than a little surprised regarding this question of whether to include zero value responses – or not, in calculating a projected salary budget. To include or not include the zero values would seem to be not unlike a familiar scenario we get into with our salary survey data extraction. That scenario is the oh-so-rare cases of management urging that certain “low data” be omitted. That kind of data “cherry-picking” seems to parallel this issue of including or excluding the zero values. It’s not like the reported values are “no responses”, which would warrant exclusion. If the object is to obtain a logical sample of the likely movement of the labor market, I couldn’t imagine omitting the zeros, since those anticipated employer actions would ultimately translate to the actual movement (or lack thereof) of the labor market.
And while I’m sure it was just simply an unfortunate choice of words, I did wonder about JC’s characterization that by omitting the zeroes the result was “3.1%, and while less the 3.8% of yesteryear — but it is still significantly better than 2.2%”. I guess I would have probably said, . . . it is a still significant increase over 2.2%. The use of “better”, almost seemed to suggest some sort of qualitative improvement of one value over another. While pretty minor, I noted that same thing a couple of years ago in the SBS report, with a number of instances where the section narratives contained some similar seemingly qualitative references to the data from one year being “better” or “improved”. Thankfully, those types of references have now virtually ceased, because as we know, the data is not good or bad . . . it just “is”.
Paul Weatherhead Program Manager Member Since: 5/1/2000 Comments: 467
Great discussion! It looks similar to a discussion string (Merit Budget Surveys) that just occurred on the Online Community.
And I like JC's alternatives for publishing the data. I referred to it as dispersion or variability reporting, but his simple illustration sure beats my academic terminology.
July 28, 2009 — A few weeks ago, WorldatWork published high level data from our 
JC, I think you are right on target. I think the larger point you are making is that you can not use SBS data, or macro market data in general, without understanding the underlying issues and trends. You also have to make sure your data reflects your peer group and your labor market. Macro data by itself is headline news, but it doesn't add much value beyond that when it comes making value-added, data-driven decisions about something as important as salary budget decisions.
Print this Page
Maybe its time to start reporting the numbers as a range of responses instead of a single average? For example, maybe the best way to report the numbers is like this: (using made up numbers for illustration only)
Budget Population
Change Surveyed
0%-0.9% 30%
1%-1.9% 10%
2%-2.9% 10%
3%-3.9% 30%
4%-4.9% 15%
5%-5.9% 5%
6%-6.9% 0%