Worker Comp Calculations 2010

In 2009 Free Geek was audited by our workers compensation provider. As a result some changes to how this work took effect at the beginning of 2010.

• In 2009 we had three categories of jobs as outlined here:

ID	Code	Description
5	7380 07	Drivers/Chauffrs/Messengers & Helpr
7	8810 03	Office Clerical
8	5191 01	Computer Dev-Instl/Inspec/Ser/Repr

• In 2010 category 8 split into two different categories and the old category 8 no longer applied, so we had four categories:

ID	Code	Description
4	3574 02	Computing/Office Machine Mfg-Noc
5	7380 07	Drivers/Chauffrs/Messengers & Helpr
6	8264 10	Recycling Operation
7	8810 03	Office Clerical

In all cases each of those categories is tied to a group of jobs that people do. At the outset of 2010 we went through the jobs associated with category 8 and divided them into jobs that were now associated with categories 4 or 6 depending.

Prior to the 2009 audit we were calculating how much payroll money was spent in each category by:

1. Looking at the jobs scheduled for the time period.
2. Comparing the total hours scheduled versus the total hours reported and arriving at a number of unscheduled hours worked.
3. Interviewing staff members about what work they did in their unscheduled time.
4. Classifying scheduled hours based on jobs scheduled.
5. Classifying unscheduled hours based on the interviews.

At the audit we were told we would need to start tracking the categories based on daily reporting by individuals of what jobs were worked. (Alternatively we could count all jobs in the highest rated category.) So we implemented a new hours tracking system in the database which went online on December 1, 2009.

Prior to January 24, 2010 all workers were paid the same (back for several years anyway), so a simple division based on hours in each category could be used to calculate how much salary money was spent in each category. As of January 24, 2010 there were three different pay scales introduced based on worker classification. Therefore, after that time hours needed to be grouped by both worker comp category and worker classification.

To complicate things further, some workers moved between categories at a few points between January 24, 2010 and the end of the next worker comp period on May 31, 2010, so we could not assume that a worker that was in one job class at the beginning of the period was in the same class at the end of it.

Method used for reporting 2010 numbers

When reporting the numbers for June 1, 2009 through May 31, 2010, then, a fairly complicated algorithm needed to be employed:

• For up to November 30, 2009 we used the old worker comp categories (5, 7 and 8).
  • Hours were calculated as outlined below and these percentages were applied to salary dollars for each period, regardless of worker class.
• For December 2009 we also used the old worker comp categories (5, 7 and 8).
  • Calculations were based on logged hours and these percentages were applied to salary dollars for each period, regardless of worker class.
• For January 1-24, 2010 we used the new worker comp categories (4, 5, 6 and 7).
  • Calculations were based on logged hours and these percentages were applied to salary dollars for each period, regardless of worker class.
• For January 25, 2010 forward we used the new worker comp categories (4, 5, 6 and 7).
  • Calculations were based on logged hours and these percentages were applied to salary dollars for each period, taking into account of worker class.

1. We reviewed the scheduled hours by their worker comp categories for the period of June 1, 2009 to November 30, 2009 (period A).
   • During this period we did not have a daily hours recording by job in place. This gave us minimum numbers of hours in each category but not a full count of the hours.
2. We reviewed actual logged hours by their worker comp categories for the period of December 1, 2009 to May 31, 2010 (period B).
   • During this period we did have a daily hours recording by job in place. This gave us actual reported numbers of hours in each category.
3. We calculated the percentage in each category for period B and then applied it to the total reported hours for period A.
   • This gave us numbers that we could compare to the scheduled hours and we could determine if they were reasonable changes or not. The most significant changes were:
     • the "Office Clerical" hours went up, which makes sense since we don't routinely schedule many tasks that happen in the office, for example email correspondence and such.
     • the "Drivers/Chauffrs/Messengers & Helpr" category dropped slightly, which also makes sense, since we schedule pickups but they do not always happen.
4. For January 1-24 (period C) calculations were based on logged hours and these percentages were applied to salary dollars for each period, regardless of worker class.
5. We identified period boundaries where workers changed classifications starting on January 24, 2010. This was in anticipation of needing to apply different amounts of salary dollars based on the pay scale for the worker class. This created three periods:
   • January 25 - February 7 (period D)
   • February 8 - April 18 (period E)
   • April 19 - May 31 (period F)
6. For each of those periods, we listed out hours logged by worker. We grouped the workers by worker classification, and then subtotaled hours by both worker comp category and worker classification.
7. We combined the subtotals from periods D, E and F into one set of hours subtotals for January 25 to May 31. Each of these hours subtotals was multiplied by the regular wage for the corresponding worker class to weight each properly. From this weighted amount we created percentages.
8. We then applied percentages based on these subtotals to the adjusted salary dollars for the same period.

Calculating adjusted salary amounts

(We use a paid time off (PTO) system and do not have an easy way to separate out vacation from sick time taken, so we elected to not deduct anything for pre-approved vacation time. We did deduct overtime wages in excess of straight time.)

To arrive at the salary dollar amounts:

1. Salary numbers are entered into our bookkeeping software and reconciled on a monthly basis.
2. We needed to deduct a portion of the overtime pay where overtime was paid out. (Overtime is a very small amount of our salary expenses, however. 62.9 hours of overtime were reported in the whole year, out of 40,759 total hours reported.)
   • Overtime is paid at time and a half, therefore we need to deduct half of the regular rate depending on the worker class for the worker who was paid the overtime rate.
   • Prior to January 24, 2010 all workers received $11.55 per hour, so the adjustment rate for overtime in that period was $5.78 per hour.
   • In all but one pay period (March 8-21 in period E) all overtime happened within a single job class. That job class has an hourly rate of $13.55 per hour, so the adjustment needed to be $6.78 per hour.
   • In the pay period of March 8-21 (in period E) all overtime happened within two job classes. 3.25 hours were in the $13.55 job class ($6.78 adjustment rate), and 4.60 hours were in a job class with a hourly rate of $12.55 per hour so that adjustment needed to be $6.28 per hour.
3. The salary numbers, then, were pulled from the bookkeeping records, overtime adjustments were applied, and the net (total salary minus adjustments) was used as the salary amount to report.