Collection and Ageing of Salary Survey Data

Wednesday, January 13, 2010

Once the analyst has achieved the maximum correlation between matching the organization’s positions to those used in the survey(s), data must be collected and “warehoused” in such a manner as to promote effective decision-making. There are numerous commercial tools and websites (PayScale, Salary.Com, et al) as well as consulting firms (Towers Watson, AON, et al) available to assist practitioners in sourcing, storage, and analysis of data. Depending on the organizations “technical acumen”, different organizations will need or expect different levels of technical analysis of data. Having once worked with both pipeline engineers and bankers, each had a different understanding and acceptance of “data” analysis. The goal of the analyst should be to understand, use, and be able to explain analytical methodologies in language appropriate for their respective audience.

Depending on when compensation data was collected, the data may need to be “aged” to account for changes in the wage market since its original collection. If the analyst has access to some “inflation index” such as the Consumer Price or the Producer Price Indices, one or the other could form the basis for ageing the data. Keep in mind that the Producer Price Index (PPI) measures price change from the point of the seller whereas the Consumer Price Index (CPI).measures price change as viewed by the purchaser.

CPI’s are published monthly and are available at the level for national, regional, and most major cities. The CPI is sub-divided into two broad population classes, CPI for All Urban Consumers (CPI-U) (this covers about 87% of population) and CPI for Urban Wage Earners and Clerical Workers (CPI-W) (which covers 32% of the population). The CPI is broken down into several components, including: food, energy, new and used cars/trucks, apparel, shelter, transportation, and medical care. Neither of the CPI’s includes data on living in rural non-metropolitan areas, which can present a problem for salary structures in remotely located operations. The CPI-U is the most commonly used index to “age” survey data.

Aging compensation data allows the surveyed data to be “brought forward” to the current point in time. Since employers do not all adjust workers’ salaries at some uniform time, survey data represents a sampling of compensation practices over time and is thus a lagging indicator of market wage rates. Some online survey providers automatically age data based on location Zip Codes. The analyst should always verify the ageing process should they be called upon to explain it.

Assuming that survey data available has not been automatically aged by the provider, the analyst will need to determine the rate of change from the time the data was effective to the point at which the data will be used in the decision-making process. If data is being aggregated from multiple sources all with different time frames, the analyst will need to use different ageing factors for each data sources. If the survey was published May 2008, based on compensation effective January 1, 2008 and the analyst plans to use the data for his practice effective Jan 1, 2010, then 24 calendar months will have elapsed since the data was in effect on January 1, 2008.

The analyst must now calculate the change in the CPI-U for January 1, 2008 to January 1, 2009.

December 2009 CPI-U .......... 216.33 ........ Not seasonally adjusted
January 2008 CPI-U .............. 211.08 ........ Not seasonally adjusted.
CPI Change:( ageing factor) . 2.487% ....... Percentage points.

Based on this, the analyst must age the survey reported compensation by a factor of 2.487 percent in order to align it with the expected update on Jan 1, 2009. Making a change this small could have both negative and positive employee relations aspects. On the one hand, the organization is taking action to maintain wages at a “competitive” level. On the other had, a two and half percent change will have only a nominal effect on most employees and may be perceived as an insult to some. Much will depend on how the organization chooses to go about communicating the message rather than the message itself.