Back in July 2015 I attended a meeting at RAND in Santa Monica along with a group of workers’ comp stakeholders. The meeting was to get a preliminary look at a RAND study being conducted to assess wage loss replacement under SB 863.
The outlines of the study are becoming more clear. The RAND researchers doing the study, Michael Dworsky and Seth Seabury, presented their findings to the December 9 meeting of CHSWC in Oakland.
RAND studies have often been cited by policymakers in formulating California workers’ comp laws. So studies of this type may have real consequences and aren’t just for wonks.
Neither the slides presented at the December 9 CHSWC meeting nor the research report itself have yet been uploaded onto the CHSWC website as of the date of my post.
But in the interests of keeping the wider comp community informed, I’ll provide a summary of some of the important points (based on my notes). However; first a caveat. This was a presentation lasting over an hour, chock full of graphs and statistics. So readers are advised to read the report itself carefully when it is made available to cross-check the accuracy of my notes. Due to space limitations, in this post I’ll present my notes on the presentation on the study itself rather than dissecting questions about the methodology or the conclusions.
Seabury began with a history of the effect of the various reforms. Between 2000 and 2006 there was a sharp decline in lost income replacement rates.
The general state of the economy is a significant factor in wage replacement rates, and the state of the economy may affect disabled workers even more. The “Great Recession” that hit around 2008 had severe impact of California workers, creating a large unemployment spike.
Low wage replacement rates became one of the motivating factors of the 2012 reforms. SB 863 was touted by proponents as intended to improve benefit adequacy.
The study looked at how labor market outcomes for permanently disabled workers evolved between SB 899 (2004) and SB 863 (2012). RAND developed a large data set, including linking case information from the WCIS system to quarterly UI data. They settled on a methodology by matching data on injured workers to data on uninjured control workers (who worked for the same employer at the time of injury and who had similar earnings prior to injury). In doing so they developed stats on workers injured between 2005 to 2012, showing earnings losses driven by reduced employment following injury.
Dworsky posed the question of what explains the decline in earnings and return to work? Was it driven by a change in composition of the claimant population?
Apparently RAND found important differences between workers injured before and after the recession. Comparing workers injured from 2005-2007 to those injured 2008-2012, the latter tended to be those earning a lower pre-injury wage, to have had more severe injury (based on TTD duration), to be older and more likely to be female, to have a higher pre-injury job tenure, and more likely to have worked in smaller firms.
Low wage workers are said to lose a higher percentage of their earnings. RAND data studied did not include data on ethnicity or citizenship status.
So the assertion was made that the changing composition of PD caseload helps account for earnings loss trends; these compositional changes seem to have accounted for some of the worsening wage replacement outcomes. But compositional changes in the PD caseload may have also resulted from the Great Recession.
But did benefits (the study seems to looks at benefits awarded rather than benefits paid) awarded keep up with increased earnings losses? The WCIS data suggested that claims have a higher severity in recent years and that ratings had increased steadily under SB 899, especially among represented workers.
In assessing adequacy of benefits, the National Academy of Social Insurance has proposed 80% after tax wage replacement as a goal. Using a 5 year horizon to assess wage replacement is likely to underestimate total lifetime losses and overestimate wage replacement, however.
In any event, after tax wage replacement fell from 2005 to 2012.
The remainder of the presentation focused on the effects of SB 863 (the 2012 reforms).
SB 863 made changes to impairment ratings and benefits. Weekly maximum wages used to calculate PD benefits were raised, an FEC 1.4 modifier instituted, add-ons for psyche, sleep and sex were limited, and a $5,000 “Return To Work Fund” created.
Dworsky noted that from some early DEU data it appears final ratings are a bit higher for post 2013 claims. But he also noted that it is really too soon to use actual ratings for post 2013 claims in evaluating wage loss replacement.
So to focus on the effect of 2013 RAND decided to analyze how the SB 863 changes would have affected wage replacement claims if the SB 863 provisions had been in effect from 2005 to 2012.
They propose to do this by charting SB 899 statutory benefits, SB 899 statutory benefits but applying the SB 863 weekly maximums, SB 863 benefits and SB 863 plus the RTW Fund. They then project after tax wage replacement rates over the period from 2005 to 2012.
If I understand their conclusion correctly, SB 863 increased maximums and other changes have had little impact on the lowest wage earners, but the RTW fund may be important for low wage workers.
There was also discussion of the impact of the various reforms on vertical equity (do workers with more severe losses receive higher benefits?) vs. horizontal equity (comparing recovery among similar injuries).
SB 863 doesn’t appear to have substantial affect on vertical equity and is unlikely to affect horizontal equity across types of conditions.
At the end of the presentation Dworsky was candid about the limits of the study. The focus is on statutory benefits, not benefits actually paid out. Rating trends may change as parties evolve and adopt to the new law. And the report does not actually focus on the effect of Almaraz/Guzman ratings, though it offers some speculations and highlights some “emerging issues”.
The conclusion reached by RAND is that changes to benefits under SB 863 will lead to more adequate disability benefits than under SB 899.
Whether long term data will back up that conclusion isn’t clear, but for the moment the RAND conclusion will be trumpeted by the various administrators and stakeholders who were instrumental in SB 863.
The presentation was followed by some robust questioning from CHSWC Commissioners Kessler and Bloch, both of who seemed very concerned about wage replacement for low wage workers.
At the end of the meeting it was agreed that the concerns raised by some of the CHSWC commissioners would be included in the report or a document to be drafted and attached to the report.
So stay tuned.