From: Knapp, Allen (VDH) Sent: Friday, September 02, 2011 1:33 PM To: Bowles, James (VDH); Roadcap, Dwayne (VDH); Bolling, Patrick (VDH) Subject: Re: Privatization study Guys, This came up in my conversation with Ben Berteau of Infiltrator. Basically he wondered why we weren't fighting this. I explained that we don't take sides when it comes to legislation...that we provide information, etc. Then it occurred to that if we had concerns about this legislation, we would express them. Then I thought, the only basis we could have for concern would be public health or environment...then I wondered if there might be any data showing any difference betwEen private work and vdh work. Makes sense on some level- going into the AOSE program we said ther was no data to show that private evaluations are any riskier than public. Now where are we after more than decade? Allen. From: Bowles, James (VDH) Sent: Friday, September 02, 2011 08:52 AM To: Roadcap, Dwayne (VDH); Knapp, Allen (VDH); Bolling, Patrick (VDH) Subject: RE: Privatization study I hadn’t thought about trying to compare the failure rate of permits issued by VDH to the failure rate of permits issued by OSEs/PEs. I think it can be done, at least on a superficial level, but even that would take some time because the teasing the data out would be complicated, given the current structure of the data base. Once you identify the systems that have failed, it’s not always easy to categorize whether it’s a VDH permit or a private sector permit. For example, if an EHS issues a construction permit based on an OSE certification letter, is that an EHS permit? What about if an EHS does a site evaluation and then a PE designs the system? Failing to stratify the comparison for these differences could produce very misleading results. Then there’s the problem that VDH doesn’t design permits for alternative system. So, in the end, we might be left with only a relatively small number of systems for comparison: conventional systems for which the original site evaluation and design were both done either by an OSE or by an EHS. If we don’t have a large enough population of data, then the results may not be particularly meaningful. And I don’t think that we want to just do a “preliminary analysis” because once it’s done, the data is going to be out there, and it’s going to be difficult or impossible to control the use of bad (or at least incomplete, insufficient) data to support a position that the data, on the surface, seems to support. The bottom line, IMHO, is that we can do this but it will take some planning and some time. So we need to decide what we won’t do if we decide to pursue this. But it would be interesting, and I’d be willing to pursue it if you have the time to spend helping design the study. Jim