TPA Administrative Costs: A Review of 2011 Results - Sherlock ...

Transcript 

TPA Administrative Costs: 

A Review of 2011 Results 

September 5, 2012 

Douglas B. Sherlock, CFA 

sherlock@sherlockco.com 

(215) 628-2289 

 

Welcome, and thank you for attending our third annual web conference on the 

administrative costs of TPAs. Today we’ll discuss the summary as published in the 

September Plan Management Navigator. 

 

In this presentation I’ll present some background on the study, discuss the expense 

levels and provide a comparison with comparable products of forty-two health plans in 

other Sherlock Company universes. 

 

Sherlock Company is focused on performance benchmarking of health benefit 

organizations. We think that the combined effects of the weak economy and regulatory 

flux make this effort especially timely. 

We are the leader in performance benchmarking for health benefit organizations, 

having done this for fifteen consecutive years. Sixty-two organizations participated this 

year, collectively serving nearly 50 million Americans with comprehensive products. 

While we have literally thousands of analyses in the TPA benchmark reports, the 

content reflects the preferences of the panelists, which keep us to a high insight-to-effort 

Page 1 of 10 

C OPYRIGHT © 2012 S HERLOCK C OMPANY. ALL R IGHTS R ESERVED.

atio. We are great believers in the wisdom of markets so we are very grateful that our 

benchmarks enjoy broad acceptance. For instance, organizations serving 143 million 

Americans with health coverage are users of our 2011 editions of our benchmarks. 

 

I am going to touch on the numbers shortly but before I do, I wanted to discuss the 

universe itself. There were eight participants. While this is a small universe, it is likely 

an elite one. 

The panel TPAs served 3.1 million members through a variety of different products. 

The average revenues of the TPAs was $33.9 million, and we estimate that is about three 

times the size of the average TPA. With, on average, 181,000 medical members, this is 

also considerably higher than average. 

While 56% of the members were Medical, Medical also comprised on average 85% of 

the panel’s revenues. Other products included Dental, FSA Admin., Pharmacy, Short- 

Term Disability and Student Health. 

 

Developing useful benchmarks for the TPA industry has unique challenges. I want to 

first discuss them, and then tell you how we have adapted to the measurement 

problems. 

The first challenge is that the universe is certainly finite. Based on some figures from the 

Society for Benefit Plan Administration, in 2010 there were 281 such organizations. I 

understand that there are 270 now. This excludes specialty TPAs, ASOs run by insurers 

(though we’ll discuss them later), companies doing a modest amount of TPA work on 

the side and firms simply using TPA in their descriptive materials. The SPBA believes 

that its membership comprises north of 95% of all true comprehensive TPAs. 

As you can see in this chart, we’ve estimated that the combined revenues of the 281 

TPAs are $3.1 billion and that they collectively serve the equivalent of approximately 20 

million Medical members. 



But even more interesting is the distribution of the TPAs. Of the 281, 254, or 90%, have 

revenues of less than $20 million. In this chart, 90% of the TPAs are represented in the 

lowest revenue six brackets. This highlights an important barrier to participation in our 

TPA benchmarking study. Let us suppose that all TPAs have exactly the same staffing 

ratios that our participants had – we can then estimate the people available in these 

TPAs to participate in our benchmarking study. 

For our panel, the average Finance and Accounting staffing ratio is 0.33 FTEs per ten 

thousand members, excluding the people tasked with fund accounting. If you assume a 

midpoint revenue of the middle SPBA category with revenues $5-$10 million of $7.5 

million, and a per member monthly revenue yield of about $13.00 PMPM, you can 

estimate that this bracket’s average membership equates to approximately 49,000. Given 

our Accounting function staffing ratio of 0.33, we would expect that the typical TPA in 

that bracket to have an accounting staff averaging only 1.6 FTEs! Even assuming that 

smaller TPAs have higher ratios due to economies of scale, it illustrates the difficulties 

in TPAs finding the band width for participation. Hence, our universe is limited in part 

by the resources of the TPA industry participants. 

 

The second challenge is the complexity of the products themselves. Each TPA offers a 

variety of different products and each of the products that it offers is available with a 

variety of different supplemental services. As noted before, their businesses emphasize 

Medical, but Pharmacy, Dental and Vision are also prominent. So, on its face, it is 

difficult to compare member to member since members’ products are not equal. 

Compounding this problem is that, for each of these products, customers may elect to 

purchase supplemental services. Supplemental services offered by these TPAs for an 

extra fee include Medical Management, Provider Network Management and Services, 

COBRA and HIPAA Administration and Mental Health Administration. So, even if 

TPAs offered only one product, the product itself would not be directly comparable to 

those of other similar organizations if their customers demand a different suite of 

supplemental services. 

Thus, there are literally thousands of combinations of products available to TPA 

customers that directly impact comparability of the TPAs. 



Fortunately, we’ve been able to develop solutions to the twin problems of the available 

band width for the average TPA, and the complexity of their product offerings. 

First, the band-width problem. We solved this, or rather, the TPAs solved it for us, by 

participation being skewed towards larger organizations. The average revenue for 

participating TPAs was $33 million and seven of our TPAs have revenues greater than 

$20 million. Our smallest has approximately $10 million in annual revenues. 

The seven TPAs with greater than $20 million in annual revenues are actually part of a 

rarified group. From back on slide 5, you have only 27 TPAs in those top three brackets. 

So our seven largest TPAs approaches about 25% of the total in those three brackets. 

By the way, there may be some interesting implications from this skewing. To the 

degree that size carries with it some operating advantages, it would follow that these 

organizations would demonstrate them. More importantly, our TPAs have self-selected 

to be not only larger, but also have the staffing necessary to perform the measurement 

so crucial to achieving overall cost leadership. For those reasons, plus their comparison 

with other universes that we’ll develop later, we think that these are likely elite 

performers compared with their peers. 

 

Recall the second problem that we faced, achieving apples to apples comparisons in an 

environment in which TPAs offer many different products, and each of those products 

may differ in the scope of their services. 

To solve this, we decided to finesse the challenge of comparing TPAs by focusing on the 

intrinsic comparability of each product/expense cell. In other words, while the TPAs 

are themselves not perfectly comparable, functions within each product most certainly 

are. 

This slide shows what I mean. First, the participants are invited to segment their costs 

by core and non-core services. Medical management is an example of a non-core 

service. Then we ask the participants to further segment all costs by function, with 



Group Reporting and Enrollment being examples. Finally the TPAs segment their 

functions into each product that they offer. 

This level of granularity is tedious but necessary to achieve comparability in any of the 

approximately 320 product/expense cells. This has a second advantage in that we can 

then compare each TPA to each other simply by reweighting their function costs to 

assure the exact same product mix between them. That analysis is beyond the scope of 

this presentation but, as you might imagine, that is a pretty powerful advantage of 

granularity. 

As an aside, as you might gather from this slide, we standardize costs by employees 

(sometimes called contracts), members (that is employees plus dependents) and 

revenues. The denominator always relates to the actual user of the core or supplemental 

services, as the case may be. 

As we’ll develop later, we do not use this approach only for TPAs; we use it in all of our 

other universes. This extreme segmentation has the side effect of allowing us to 

compare TPAs to similar products of other organizations. 

While actionable information requires this level of specificity, in today’s discussion 

we’ll summarize by solely reporting high-level metrics. I’ve identified the scope of them 

in pink on this slide. We’ll speak about total TPA costs and their functional 

components, shown in the far right column. And we’ll also talk about total core product 

expenses, shown in the row on the bottom of the top chart. Because this is simply a 

summary, we won’t go into the non-core services at all, or core services by product. 

 

As you might imagine, we are exceedingly careful in our data collection and reporting. 

We codified reporting in a set of Guidelines and also imbedded these Guidelines as 

comments in the survey form itself. Even though we sought consensus on these 

definitions, reporting to us almost always entailed reclassifications of their own 

expenses by the reporting TPAs. 

After the initial data submission, we identified outlying responses and pursued them to 

make sure that the submissions were an accurate representation of the actual 



performance of the TPAs. This normally entailed multiple refinements of the original 

survey submission by each TPA. 

To double check the data, we asked that it be reconciled with the audit, which the 

participants supplied to us, after having adjusted for the peer group’s reporting 

conventions. This reconciliation amplifies the effect of our check for outliers by 

magnifying differences in each product/expense cell. 

So, before I go any further, let me offer a hearty thanks to our TPA participants who 

made these benchmarks happen. I know it is a lot of work. But I suspect that you will 

find that your organization will benefit greatly from these benchmarks. And, one more 

thing I can assure you, there is a learning curve. So you will find that it will be much 

easier the next time around. 

 

And now, to the results. For ease of communication, we have summarized the 

functional costs into three clusters of expenses: Sales and Marketing, Account and 

Membership Administration and Corporate Services. The principle functions within 

these clusters are listed in Appendix A of this presentation and we will mention the 

component functions as we discuss them. 

TPAs classically report costs standardized by employees served, and this slide shows 

the values of the reporting TPAs using this approach. Core costs have a median value of 

$11.87 PEPM, and the largest portion of those costs are Account and Membership 

Administration. At $7.06, this cluster of expenses exceeds the other two clusters 

combined. No surprise here though. This is the area that one normally visualizes when 

thinking about the health benefits business: Claims, Enrollment, Customer Service and 

Information Systems are the functions included here. 

 

In thinking about these per-employee served metrics, and all of the other metrics of 

costs that are necessary to support these products, it is important to bear in mind that 

these represent product mix-weighted values. Each of the products has different cost 

values as shown in this slide. The Medical product has costs of $20.94 per employee per 

month, and it dwarfs the next highest cost product, Flexible Spending Account 



Administration, at $5.41 per employee per month. For the TPAs in our universe, fees 

attributable to the Medical product ranged from 74% to 100% of the total. 

 

For those of you participating in this web conference who work for health plans as 

opposed to TPAs, you may be more familiar with costs standardized by member 

months. This slide expresses costs standardized in this way. The median costs per 

member per month is $5.79, about half the value of the per employee value of $11.87 

shown earlier. This is not surprising since the average contract size (members per 

employee) ranges from 1.91 to 2.26 for the Medical product. Actually, the range in 

values is one of the reasons why we like the per member per month approach: because 

of that 0.35 contract size difference I just mentioned, 18% of the differences between 

TPAs using the per employee served values could stem from differences in contract 

size! Put a different way, if costs relate to people’s needs for their claims to be processed 

or their phone calls answered, then the membership relates well to those TPA activities. 

Again, Account and Membership Administration comprise the lion’s share of Core 

Service costs, at $3.41 PMPM. Sales and Marketing costs which, by the way, includes 

Employer Group Reporting as well as Sales, Marketing and Advertising and Promotion 

was the third largest component of the costs at $0.88 PMPM. Broker commissions are 

not included, as they are treated as a pass-through by TPAs. Their independent audits 

tend to corroborate this treatment. 

 

Again, it is important to note that costs vary greatly by product, as shown in this slide. 

The Medical product is again the high cost product at $10.26 PMPM. But while the 

PMPM value of the Medical product is 49% of its per-employee value, the Vision 

product, at $0.28 PMPM, is 41% of its per-employee value. So, not only the does the size 

of the contract make a difference in the benchmarks for TPAs products as a whole but 

how the size of the contract varies within each of the products also makes a difference. 

In my view, this is an especially important slide in that it shows the per member costs of 

the Medical product, the central product for this universe, in a way that can be 

compared to the metrics reported by other health benefit organizations such as health 

plans. 



Let me reemphasize that the values in this chart are only for core services like 

Enrollment, Claims and so forth. They do not include Medical Management, Provider 

Service and other supplemental (or non-core) services that may be elected in addition to 

the core services required by the employers. We calculate the non-core costs separately 

and denominate their costs by the members or employees who use them. For instance, 

the non-core Medical Management service costs for the Medical product is $1.56 PMPM 

or $2.83 PEPM, where those medical employees or members are users of these non-core 

services. 

 

The costs of all organizations, including TPAs, are sometimes measured as a percent of 

revenues. Core costs for these TPAs have a median value of 87.6% of fee revenue. Again 

Account and Membership Administration costs dominate TPAs’ cost structure, 

comprising nearly one half of revenues, at 49.6%. 

Note that Corporate Services costs, which include Finance and Accounting, Legal, HR, 

Corporate Executive and Governance, have a median of 21.7% of fees. As you may 

know, Sherlock Company believes that economies of scale in health benefit 

organizations are few and modest. When we compare TPAs Corporate Service levels to 

the approximately 21-27% of ASO fees for similar Corporate Service activities of much 

larger Independent/Provider-Sponsored or Blue plans, I think we have another mosaic 

piece supporting our belief in modest scalability. After all, if there were competitively 

significant economies of scale, this is where you would find them. Success in 

administrative cost management for health benefit organizations is thus all about the 

execution. 

Speaking of percents, these TPAs showed impressive profits in 2011 in their core 

services. The median margin is 12.4%, with 25 th percentile value of 5.8% and a 75 th 

percentile value of 23.1%. We don’t publish trend data for this universe but I would be 

remiss if I did not add that the TPAs that participated for their 2010 years had median 

margins of 1.6%. 

The non-core services offered by these TPAs enhance their profitability slightly above 

this level to 13.0%. By the way, note that 100% minus the 87.6% cost equals the 12.4% 

profit margin. 



As with the other cost metrics, operating costs as a percent of revenues varies for the 

core products offered by these TPAs. Student Health Administration is the low cost 

product with costs of 37.7% of fee revenues, followed by Pharmacy at 49.1%. FSA 

Administration is the high cost product 110.9% followed by Vision administration at 

102.4%. Operating costs are 93.5% of the Medical product. You can get a rough sense of 

the profitability of each product by subtracting the median costs from 100%, similar to 

the previous slide. 

 

I earlier said that I suspected that the TPAs that participated in our benchmarks are 

relatively elite performers because of their operating characteristics. This slide shows 

their actual performance also appears superior. When we compare the costs of our 

TPAs with similar products offered by Blue and Independent / Provider- Sponsored 

plans we find that the TPAs’ costs are usually lower than those of the other health 

benefit organizations. Median costs for the Medical product was $10.26 for our TPAs, 

$16.76 for the Blues and $12.18 for the Independent / Provider-Sponsored plans. In fact, 

there appears to be little overlap between the TPAs and their Blue competitors as the 

25 th percentile value for Blues is greater than the 75 th percentile value. 

So, if these TPAs comprise a sample that is statistically small, they nevertheless are 

organizations that appear elite, function for function. By contrast, our sample of Blue 

plans, representing two-thirds of continental US primary licensees, is likely 

representative of the Blue universe as a whole. It may also be that some services are 

provided by Blue and IPS plans that are not provided by TPAs, but somehow we’ve 

failed to isolate them. So while it may not be possible to precisely extrapolate across the 

different sectors of health benefit organizations, this chart is further evidence that 

illustrate that smaller organizations, which the TPAs are compared with the other peer 

groups, are not precluded from low costs. 

In this comparison, we matched as best as we were able, the functions between the 

universes. For instance, the costs of broker commissions were excluded from the costs 

of all of the peer groups, as is the TPA reporting convention. Similarly, we excluded 

Medical Management and other services like that TPAs and their customers consider 

non-core. 



When we look at core costs for their products as a whole, per employee monthly costs 

are $11.87 and per member costs are $5.79. More meaningfully, total core monthly 

administrative expenses for the Medical product are $20.94 per employee and $10.26 

per member. Costs run 88% of revenues overall, and 94% for their medical products. 

TPAs have profits that were higher this year than during last year’s survey, and their 

supplemental services increase their overall returns. 

Function for function, these selected TPAs have lower costs than are typically found in 

competitor organizations. By this comparison, these TPAs are superior performers. 

Before I go to questions, I’d like to show you’re the appendix slides. We’ll place these 

slides, along with the transcript, on our website in the next few minutes. 

 

Now I would like to open this for questions about the results of the benchmarking 

study. 

Questions 

I want to again thank you for your participation in this web conference. More in depth 

and actionable information is available in the benchmarking study itself, which anyone 

can license. Please contact me directly if you are considering licensing these materials. 

Over the next month or so, we will have similar web conferences on the results of the 

Medicare and Medicaid plans. We hope that you will consider participating in these 

web conferences as well. Transcripts, slides and summary results on the Blue Cross Blue 

Shield and Independent/Provider-Sponsored benchmarks are already posted on our 

website. 

Finally, I want to close by once again thanking all of you who participated in this study 

for your efforts. Your participation not only enhances your own firm’s performance but 

also raises the bar for all other health benefit organizations. 

This is Doug Sherlock of Sherlock Company.

TPA Administrative Costs: A Review of 2011 Results - Sherlock ...

Create successful ePaper yourself

Delete template?

Save as template?