What software do you use for mid-size (200 - 1,000) plans? How well does it work for you? We currently use Datair and are looking to upgrade our software and would like to get some feedback from users of other products. Thanks.

We use Proval, which is a Wintech system. It is a very good system w/ good support from Wintech. It's fully iterative and gives you the ability to do projections w/ ease.

It is quite a jump in cost from Datair, but you get what you pay for.

Many thanks for the info. Actually, the two systems we are considering most seriously are ProVal and ASC. The bean counters are most interested in SMART, which I've used in the past, and BLAZE because they cost less; and nobody except the actuary is looking seriously at Lynchval because of cost. If anybody has any experience with any of these systems I would really appreciate hearing from you.

Well, I'll take the opportunity to give Max "Smart" Rosenberg a plug, as I used his system and services for a long time in a different life. But Max I think calls his system a small plan system. At least he used to.

I've used Blaze but much prefer Smart/INTAC.

In my current life we use what has now morphed into WYSTAR, which I believe originated as a Lynchval clone. We consider this to be excellent for DB plan valuations, but not so great for proposals or testing. It is clearly designed for larger plans.

Thanks, Andy. In a former life I too used SMART and found, like you, it worked just fine for small plans.

In ancient history (circa 1975), Joe Lynch worked for ACTI, the predecessor of Wystar. Lynchval is the clone. I was a client of ACTI in the 70's.

At this size plan (200-1,000), are you dealing with layered multiple benefit formulas? Many of these plans have grandfathered benefits, combo cash-balance & regular formulas, and different divisions with different vesting, past service credits, and retirement ages.

If you are looking seriously at ASC, ask about their ability to run multi-decrement valuations with multiple benefit formulas, and to check their sample life reports. I don't think they have the same capacities in these areas as Wystar, Lynchval, or Proval, or even Relius. I understand that they have to custom program any unusual plans.

For what it's worth, tell Lynchval why their prices are so bad. Also, talk with Lorraine Dorsa about her attempts to get them to be more efficient and more competitive.

For myself, I decided to write my own custom programs for clients with complex formulas. But I would use one of the commercial packages if I had more volume in this size range plan.

I used ASC a few years back and would agree with the comment that ASC needs a lot of customization and overrides for what they are claim are unique benefit formulas and designs. It seemed that 80% of our benefit formulas (small DB design) fell into the "unique" category, even though I didn't think we were doing anything that unique or special. Perhaps they've improved since then though, but I'd definitely run some demos using various types of formulas and funding methods that you commonly use and make sure you don't have to override a lot of things.

Thanks very much for the info about ASC, we will definitely ask. Way back in the really dark days I used ACTECH, which had been designed to compete with ACTI; highly user-customizable Fortran, really bare bones client reporting. Is Wystar is something like this? I gather ProVal is a very good actuarial package, can anyone comment on its reporting and data import/export features?

WYSTAR is ACTI, acquired and renamed on occassion.

Proval has very good import/export capabilities into Excel. We generally reconcile the data in Excel, dump it into Proval for the runs and dump the output back into Excel to create our reports/statements.

Proval is a very good "valuation" program. It is not really designed to do produce valuation reports, benefit statements, participant listings and/or benefit calculations (although it will do all of these things). Most of the users take the output and create their own customized reports. It also isn't the best for "small" plans. It's fully iterative and handles mult-decrements so telling it you just want to do something without pre-retirement mortality can tend to confuse it. Also, it still can not do an end-of-year valuation correctly.

I think it's a great system, but it really isn't designed to replace Datair. We still do most of our small tax-shelter stuff in Excel or Datair, even though we have Proval for the "real" retirement plans.

I will second Effen's comments. Although, once our valuation report was created in Excel, it became easy to dump Proval's output into our spreadsheet. Also, performing an EOY valuation is an easy workaround, but one we struggled with until we figured it out. It would be very good for 200-1000 life plans.

Many thanks for the info. However, I am intrigued by Effen's comment that ProVal gets confused on small plans with no pre-retirement mortality. Could you possibly elaborate on that?

Proval (and I assume Lynchval and other iterative systems) expects three mortality tables (healthy actives, healthy retirees & disabled participants). The concept that someone can't die while active "does not compute".

The idea of pre/post retirement comes out of simplified programs and small plans where the probability of death at younger ages isn't material. You can work around it, but why? Proval is designed to handle plans with tens of thousands of lives. Yes it can do a one life plan also, but getting Proval to replace Datair is kind of like replacing your Civic with a Hummer. You will have a lot more power and capabilities and it may take a while to figure out just how to use it.

If your db practice is growing and your getting bigger plans you will absolutely need a system more sophisticated than Datair. It's just not designed to handle the bigger plans.

Perhaps the issue is not pre- and post-retirement mortality (or interest rates). Proval handles that issue pre- and post-decrement.

I am curious about the pricing of ProVal, and its comparison to Wystar, Lynchval or other large plan software. How competitive are they?

Do they have monthly minimum fees? What initial fees and training expenses are involved? Do they charge multi-user licenses? How do they handle software security to assure compliance with their licensing agreement, such as hardawre locks, monthly codes, license disks, etc. ?

What level of training and experience is needed to be competent using their system? Does it need an actuarial student or senior DB administrator to get anything done?

Winklevoss
WinTech Sales: (203) 861-5514 Note: ask to speak with Joe Gilbert


Pricing is comparabile to Lynchval last time I looked, maybe a little less.

Do they have monthly minimum fees? they bill you each month based on how many keys you have and what type of keys your using (pension, medical, data, etc...)

What initial fees and training expenses are involved? You will need to ask them but I found them to be VERY fair

Do they charge multi-user licenses? Not sure I understand the question, but they charge you per user. If you only have one license, than only one person can use it at a time. Everyone can have access through the network, but the "key" restricts the number of users. They also have physical keys if you aren't networked.

How do they handle software security to assure compliance with their licensing agreement, such as hardawre locks, monthly codes, license disks, etc. ? better ask them.

In an earlier post SoCalActuary mentioned Relius. We've been under a lot of pressure to use Relius DB because we have their 401(k) and document software. We tried it and found that it is still a work in progress - it won't do multi-decrements, offset formulas and some funding methods. They claim that it will now handle multi-formulae. Given how long they've been around, I sincerely hope ASC is not worse than that.

Is anyone currently using ASC? If so, how well does it handle both large and small plans?

I use three small plan systems regularly at different times:
ASC - has multi-decrement features but no documentation of how it worked.
Relius - full featured multi-decrement with sample lives and extensive checking
but limitations on multiple benefit formulas.
Datair - hasn't even attempted multi-decrement modeling.

In doing small plans, Datair is quickest, most complete, easiest to use.
Relius has the most features, but also the most demanding of the user.

For 100 life + plans, Relius is preferable, since they handle the actuarial aspects better than either of the others. ASC has not convinced me that they are serious in large plan valuations. Datair doesn't even pretend to.

However, ASC has better pricing because the up-front fee is lower, so the total cost of ownership is better for about 3 years, unless you already have a better deal by having existing licensing from Datair or Relius.

A fair number of Relius users went to ASC when the old dos-based Pentabs software had to be replaced. It had more of the old Pentabs look and feel, although the new ASC windows version is more like the Relius user interface. These were primarily small plan administration firms in the DB area.

I have friends at each firm, and find them each to have their strengths and weaknesses, prejudices and blind-spots. Each is a work in progress.

Datair delivers updates fastest, then ASC, then Relius.

Each can do multiple plan discrimination testing, altho Datair is the easiest to use.

If you are serious about large plans, you should seriously consider a two-system solution with a good large-plan approach and a separate small plan system.

Does anybody else out there use ACTI/Wystar? We have yet to find anybody to compare notes with, and there are no user groups, message boards or anything like that.

We just looked at a demo of ASC. It was nice, but not really mid-sized or large plan software. From what we've seen so far we seem to be learning that there is no middle of the road; there's large plan software and there's small plan software, but there doesn't seem to be anything that does all aspects of both equally well. Consequently, we're also beginning to lean towards the idea of keeping Datair for our small plans and buying one license for our large plans.

CONCLUSION - After spending a lot of time looking at demos, we just got the OK to go ahead with Wystar. We selected Wystar as our first choice because it had the best combination of features and pricing for our needs. We couldn't think up anything it couldn't handle, including our non-qualified plans (a lot of these), and if we had, we could have handled it with entry-point Fortran (it really reminds me of the old Actech system). One feature that made Wystar appealing is that they have not carved out the ability to calculate actual retirement benefits as a distinct module to be purchased separately. Given the large volume of calcs that a small number of us have to do, that mattered. Also, we want to give our clients the ability to dial up their own calcs on the web. Since we already have a web site set up for our DC plans, we didn't need the extensive set of tools offered by Lynchval and Proval. Yes, I know Wystar has had more owners than Elizabeth Taylor, but I can't see something that has worked for 35 years disappearing in a market with only 2 real competitors - Lynchval and ProVal. I've attached a spreadsheet showing a very rough summary of our reviews, including prices. Note: it is not complete, so do not take the absence of an "X" to automatically mean that the software can't handle that feature. About Wystar - you can purchase one full license of DBVAL (850), and then up to 3 licenses of DPSYS (250/mo). The security on the network monitors that no more than one person can run a val calc at any time, which seemed awfully workable to us. Thanks for all the input.Click to view attachment

The comparison is great! Thank you for the follow-up.

Glen, as I mentioned earlier, we use Wystar and we (unfortunately) have some extremely complex multi-formula plans. In fact we have many of them. We have yet to find one that the system could not handle.

We don't know anyone else that uses it, however, so we were wondering if we were going in the wrong direction. Thanks for the feedback.

Glen, great comparison and thanks for sharing. Does ASC and Relius really charge $7500 and $7630 per month in upfront fees ? That's much higher than Lynchval ($3,000 per mo.) which I always thought was at the top of the fee structure.

Is Wystar DB Val the same system known (or formerly known) as PFPC DB Val ?

yup. It used to also be ACTI. We still call it ACTI.

Andy, how's the report writer in DB Val ? Is it client ready and looks decent or do you pretty much need to dump the data into your own customized report ? THx.

We do not use it, nor do we use the valuation output format that they provide (it is fairly new and wasn't an option when we started using the system). We take an export file which is automaticaly generated and we have a pretty complex spreadsheet that interprets it and puts it into readable format, allows us to select various passes for certain purposes, then we merge into a word processing program. Ideally, the spreadsheet would be in final form but we aren't there yet because our plans have too many variations.

So the final format is cumbersome the way that we do it, but our ability to make changes instantly is great. For example, say you have a union plan with a mutlplier changing and you want to use one for pvab, one for funding, one for current liability, then you have three or four sets of assumptions we essentially generate a 3 dimensional output which we pick and choose from.

The upfront fees for Relius & ASC, along with Datair, cover the initial software licenses, manuals, minimal training, and installation expenses. They are not monthly charges.

These small plan systems (Relius may be called mid-range) have a number of other comparison items for your spreadsheet, which I will attempt to revise sometime after the rush of October 15.

We also are in process of selecting a DB system. The above information has been instructive and helpful. Thanks

what is the approx cost of datair (ball park figure) compared to say wystar or asc. Glen your comparison sheet is no longer available for download........could some one give me an estimate

Anyone done discrmination testing using Permissive Aggregation with DB-DC plans in ASC ? Does it work well and what if any outside calcs do you need to do in Excel and then dump in the system ? (hopefully none).

I use ASC for multiple plan cross testing some times. You have to be careful about the parameters and the use of multiple cases. Either use SSN's in both files to match them up, or have no overlapping employee records.

The results do not appear to handle subsidized early retirement to my satisfaction, but if they are ordinary designs, you can get it completed. I still prefer Datair for the small plan tests.

Others tell me that Larry Deutch has the better system for this purpose.

SoCal, if it doesn't include early retirement subsidies in it's calcs, how about GATT-417(e) subsidies in the MVAR when a plan offers lump sum distributions upon employmente termination ?

Those are included at the option of the user. But consider such things as subsidized early retirement, like "30 & out", which I don't see in the ASC non-discrimination test.

We use ProVal. I agree with all of the positive feedback I have read here about it.

However, I have not noticed any small plan "hiccups" or other issues. Except, maybe, the commentator meant that small plans do not need as sophisticated a system as ProVal to prepare a reasonable valuation. That may be true. On the other hand, by using the valuation program to determine the raw liabilities, you can feed the results into the cost summarization module (called Valuation Sets) and use those results to compare against Excel-generated cost summarizations, which we use as a confirmation that our cost summarizations are correct.

The attachment isn't working now for some reason-- Glen, could you possibly upload another spreadsheet into a reply message in this thread?

Dave Baker