11th Annual Wall Street Comes to Washington Conference

Conference Transcript
June 21, 2006

P R O C E E D I N G S

Paul Ginsburg: I'd like to begin the meeting now, and want to welcome you to HSC's Annual Wall Street Comes to Washington Conference. The purpose of this conference is to give to the Washington health policy community better insights into market developments that are relevant to health policy. We will be discussing market developments and their implications for people's health care. This is a core activity of HSC.

Today we have an opportunity to tape a different course of information, equity and bond analysts on this topic, than we're used to hearing on some of these topics. Equity analysts advise investors about which publicly traded companies will do well and which ones will not, and bond analysts advise on the likelihood of debt repayments.

Good analysts, the kind that you see on this panel, develop a thorough understanding of the markets that the companies that they follow operate in. And they also follow public policy which often has important implications for these companies. Some analysts work for brokerage companies and advise the clients of those firms. Others work for institutional investors such as mutual funds, pension funds, or hedge funds. And we have an analyst from one of the companies whose business is to rate hospital debt who will be participating on the panel on provider issues.

This is the way that we don't lose perspective on not-for-profit hospitals which are not covered by extensively at all on Wall Street. This is an opportunity for the equity and bond analysts to take a break from their day jobs of assessing the outlook for profitability or solvency of the companies and bring their understanding of market forces to bear on questions that those involved in health policy have on their minds. We also include on each panel Washington-based health policy analysts, and they have made valuable contributions to these sessions by tying the market developments more closely to the health policy issues.

The format we're going to use is the same as we did last year. This is going to be a roundtable discussion of a series of questions that I'm going to ask, and I have shared these questions with the panelists in advance. We are going to have two sessions, each with a separate panel. The first one will cover health care costs and premium trends, and various issues connected with health insurance including the new Medicare prescription drug coverage. The second panel will cover hospital, physician, and pharmaceutical issues.

There will be opportunities for audience questions after each. There are question cards in your packets. Please fill them out and give them to an HSC staff member. Or you can go to the microphones and ask questions. We will take them from both sources. Please note, though, that the analysts are not permitted by regulations that affect them to answer questions here about the outlook for specific companies. The only time you will hear a company mentioned is if one of them is pointing to it as an example of a phenomenon that they are talking about in the insurance or hospital industries.

I want to thank the Robert Wood Johnson Foundation who funds this conference and is the principal funder of HSC. And thank the folks at kaisernetwork.org who are going to be webcasting this conference, and the webcast will be after noon tomorrow at kaisernetwork.org. HSC will post a transcript of the conference on its website by early next week.

And before you leave the conference, I would appreciate it if you could take a moment to fill out the yellow evaluation form and leave it on the registration table. Be assured that, analysts that we are, we do analyze those evaluation statements and use them to improve on the conference each year.

We have two really terrific panels this year, and they comprise several analysts who have participated in these meetings in the past, and that includes analysts who have participated in these meetings in the past. That includes Christine Arnold of Morgan Stanley, Chris McFadden of Goldman Sachs, Geoffrey Harris who is now with a hedge fund, and Bob Berenson and Bob Laszewski, the last two who comment as policy analysts rather than as equity of bond analysts. And we also have several analysts who are new to this conference, Matthew Borsch of Goldman Sachs, John Wells from Fitch Ratings, and Doug Simpson from Merrill Lynch. We are going to begin our first questions on the new Medicare Prescription Drug Benefit.

First I would like to ask the panelists about your assessment of the results of the 2006 enrollment in these Part B plans; any comments they have on what about beneficiary take-up, and has there been adverse selection? Which benefit designs are most appealing? And also to explain why the average premium came in so much lower than had been projected before this started.

Who would like to begin? Christine?

Christine Arnold: I will start with the take-up. The take-up was much better than we had anticipated. We had expected about 19.9 million Medicare beneficiaries would take up the benefit and be members of PDPs, benefiting PBMs and managed care companies.

What I am not including there is the VA, I am not including federal employees, and I am not including employers who take the subsidy. Because guess what? They had drug coverage before this thing happened, so there is really no change in the world for those people. Excluding those people, 19.9 million people was our projection, and we are at 24 million right now on a comparable basis. The total is about 38 million, so only about 4 million seniors and other Medicare beneficiaries are not represented, not having credible drug coverage.

All the data we have seen suggests that they have made a deliberate decision, that they are either healthy seniors or that they somehow did not take this step and actually sign up. The penetration has been better than expected. Because the penetration has been with the lower benefit designs and the premium has come in so much lower, at $24 per member on average, and this time last year we were looking for an average premium of $37. So $24 on average is the premium, $37 is what we were looking for.

Guess what? Seniors signed up for cheaper plans, and those plans are pretty bare-bones. The selection I think was pretty positive. We got the healthy seniors in. Only 4 million did not show up. We got a lot of seniors in like my dad who does not use any drugs, a very healthy guy, and those people are benefiting from the fact that they will have credible coverage going forward, and the key question is what is going to happen to premiums in the future over time. But that is our view of the world.

Matthew Borsch: Maybe I could just comment on taking up what Christine had mentioned, a projection of 19.9 million which I think that included the MAPD.

Christine Arnold: Yes.

Matthew Borsch: We actually initially had a fairly high projection of 28 to 20 million, excluding the MAPD figure. So it actually came in pretty close to our original projection.

That said, we went through a period late last year and early this year where we were really questioning whether our projection made sense. Bottom line, the take-up has been surprisingly strong. In terms of the benefit design and adverse selection just to make one comment here, for the program overall just given the broad participation, it does not look like adverse selection was necessarily an issue per se.

However, what remains to be seen is where there may be adverse selection between the participating plans. I am particularly interested to see how that is going to shake out between the low-priced plans like Humana's, and Humana management has indicated that they believe they are drawing in some of the seniors who do not anticipate using a lot of prescription drugs, but want to be in the plan and not get hit with the late enrollment penalty, versus the more expensive plans that are more likely to attract seniors that expect to fully utilize the benefit. But the jury is out on that question.

Douglas Simpson: Not to belabor the issue, but I would echo the comments of Matt and Christine about the take-up. It has been very robust. Seniors definitely seem to be interested certainly in some of the lower-cost PDP plans and MA-lined PD plans. I put my parents on MAPD plans and was a little bit surprised at what they were able to get for their dollar, and they seemed very pleased after the 4-1/2 hours on the Internet it took to sign them up.

(Laughter.)

Douglas Simpson: And I appreciated the fact that halfway through it there was a little block that said if you are not comfortable with computers, click this box. I am not sure how you would actually find that.

(Laughter.)

Douglas Simpson: I think we are early in the process. It is 6 months out since the launch. Thus far I think on the issue of adverse selection, this is where I'm going here, it is a little bit early to tell, but so far the utilization trends I think if anything look a little bit light.

Most of the seniors are going to start hitting the donut hole towards the end of this summer, so that will be interesting. I am sure there will be press articles around that as someone who was previously getting Lipitor for $6 walks into the pharmacy and you are in the donut hole, and it is now $80. So that will be something kind of interesting to watch.

With the timeline having expired, again, going to the issue of adverse selection, it will be curious to see how that plays out politically. One could argue that it makes sense to extend that to the end of the year and just increase participation and tape those 4 million remaining seniors that Christine referenced in her comments, so we will continue to watch that as another item.

Paul Ginsburg: Bob, I know you have a different attitude about adverse selection.

Robert Laszewski: Yes, I think it is way too early to know anything, and I have seen a lot of hard data coming from the plans. I think the first thing that is important to understand is be careful at looking at some of the things you are hearing, some of the things that analysts are being told in the conference calls and so forth.

When the plans say they are coming in at projected levels, they are really using a lot of plugged numbers yet. The data is very difficult to assess at this point. One of the things you do not hear about is there are a lot of plans out there with the dual-eligibles. I know of one plan where CMS is saying they have 200,000 more dual-eligibles than the plan knows about, and this is repeating itself from plan to plan. We have had millions of people signing up toward the end and the problem with doing an assessment of benefit ratios when you have number of people coming in is you book the premium for those people for a month, but maybe they have 2 months' worth of medicine in their cabinet and it may be a while before they actually go use the plan.

The other thing with these numbers when you look at 24 million signing up, this 5.8 million people that the administration puts an asterisk on and says they are covered troubles me, because we do not know whether they are covered or not or how well they are covered. And most importantly, I think a lot of those 5.8 million people signed up for Part D because they were told that if they did not sign up they would be paying penalties and they did not get the letter saying you have credible coverage. So I think a number of those people have probably signed up and are being double-counted. The take-up rate really varies with that 5.8 million. It is anywhere from 60 percent to 75 percent of the people who did not have coverage on December 31, depending on how that 5.8 million shakes out.

In the benefits business which is where I spent most of my career, you figure you need 70 to 80 percent of the people to get a good cross-section. So we are either at 70 percent or we are close to that, and in terms of adverse selection, we are probably close to the number. But the flip side of that is, the premiums have come in far lower than we expected that they would.

One of the news events that I think a lot of people missed on Part D was when the administration was running around congratulating themselves on the fact that the insurance companies bid 10 percent less than they thought they would bid for the dual-eligibles.

The old underwriter in me said they came in on an existing piece of business 10 percent lower than what it was supposed to cost. How did they do that? And what justifies coming in 10 percent lower? You also have a plan that looks very much like the federal employee plan of which many of you are aware, and you know what you do when it comes time to sign up for the federal employee plan. You know what your particular health care utilization needs are and you find the optimal plan, the best benefit for the lowest price.

I think that there are a whole series of factors going on here, not the least of which is that the data is impossible to evaluate. We will not really know how this came out until probably early next year. The health plans have people signing up and getting on the rolls right through the month of June. July is going to be the first month where you theoretically know who you are covering, and then did the people who signed up in April, May, and June have a bunch of medicine in their cabinets and have not used the plan yet?

The first time you are going to know what the claims side really looks like may be September or October, and it is going to be late in the year before you can really balance premium and claims. Then you are going to have the donut hole for some and not for others. Everybody talks about the donut hole which the plans are really counting on for claims to stop, except the people who signed up in May are not going to hit the donut hole because they signed up in May.

So you get the picture. I looked at real live data from a real live plan that the analysts do not get to see, and through the first quarter they were showing an 113 percent benefit ratio. You do not have to be an actuary to know that that is not good. They are counting on the donut hole and the risk adjusters kicking in, and the rick adjusters are the big factor in this whole thing.

The bottom line is nobody has any idea, which really makes it difficult now to price next year. So do not ask me what the prices are going to look like next year. As one executive who actually does the numbers told me the other day, I said I am not going to quote you on any of this stuff. How are you going to price next year? He says, crap shoot. I don't know. So it is going to be interesting.

Christine Arnold: I agree with a lot of the points that Bob made about not having transparency, but I would point out three things.

First, 75 to 80 percent of costs that come in 2-1/2 percent higher than you anticipated are borne by the federal government, so we have this thing called the risk corridor. By our estimates, costs have to be 19 percent higher than you anticipated. If you bid an 85 percent loss ratio and a 5 percent margin, 19 percent higher than you anticipated in order to get to break even from a 5 percent margin.

So we have the reinsurance, we have the risk corridors which what I have just referenced, and we have the fact that for dual-eligibles you are being paid for the risk profile of those members, and I do think those factors do mitigate the fact that we are going to have a lot of plan-to-plan reconciliations in the third and fourth quarters, and state-to-plan, because here is another issue.

The states did not turn off their drug cards. When this whole thing was a disaster in January for a month or so with just a flood of seniors in, the states did not turn off their

state Medicaid cards. So you could wander in as a dual-eligible, you just handed them the same card as you handed them for the last 5 years, right?

Here's my Medicaid card, and that would work. The states for a bunch of these, so we are going to have that reconciliation where the states want that money back because the plans have been paid premiums. So there are going to be a bunch of adjustments, number one.

The risk corridors and the reinsurance really do put training wheels on the program from an earnings perspective. And also keep in mind, point three, that in the seasonality for this business, it was intended to be unprofitable in the first quarter. Every company in my coverage universe except Wellcare reported a loss in the first quarter for PDP, and that is the way the benefit was structured. It is a second-half earnings story from a stock perspective. And I do agree that there is a lot that we do not know, but the point is that the government put on all these training wheels that I think mitigate the impact on earnings. I don't know what my colleagues feel here.

Douglas Simpson: I would add, echoing again some of the comments on the reinsurance and then the risk corridors, it's hard to really screw this thing up from a financial perspective because there is so much cushion, and it is hard to make a ton of money, too.

So it is interesting for me, for all the discussion and air time that Part B gets, it does not really move the needle for most of the companies. It just doesn't. One of the companies we cover had an analyst day in November, and roughly 2 cents out of a $3.50 number is represented by Part B. They must have gotten 25 questions during the analyst day about Part B, and yet nobody asked them what their paperclip spending was, and it is probably about the same magnitude.

It gets a lot of air time and it makes a lot of interesting press stories, but at the end of the day it does not really factor into a lot of these companies' bottom line in a big way.

Paul Ginsburg: I think that is a good point, and it may be a good occasion to turn the discussion. I do not think that this audience about how the companies come out except when it shows up, either for government obligations or what beneficiaries are going to have to--and that is what I would like to focus on.

What does it look like next year or the year after for the federal government and the beneficiaries?

Robert Laszewski: That is exactly the point. This is I think the third year I have done this, and I keep beating on the analysts to think about next year, and not just where managed care is in the next quarter.

What happens when the risk corridors are delved into, what happens when the federal government makes up for all of the underpricing, is that a good long-term strategy from a business perspective? Because government has a contract to do that today, and we are sort of on auto pilot for the second year as well. But if in fact these risk corridors bail everybody out, then what happens to the Part D policies, a Congress that is facing enormous budgets, a Congress that is already somewhat thinking twice about passing this thing in the first place and Democrats railing against the giveaways to the managed care industry?

So if you do get into those risk corridors big time and there is a significant bailout and the health plans have to provide 20 to 30 percent rate increases in the third year to catch up for all of this, you have a policy mess on your hands, and you do not have a good long-term business strategy either.

Now you have 20 million customers and it is sort of like the 1997 BBA all over again, except this time there are 20 million customers rather than 5 million customers in Medicare Advantage. So I think there is a real issue in terms of the long-term policy and the sustainability of this program.

The Medicare trustees have already said that the trend rate on Part D is about 11 percent. If all things go okay, this thing has an 11 percent trend rate in it. If it does not go okay, then start multiplying that. That is the longer-term policy issue, and it is the longer-term business issue that I think the analysts have to be asking: Company X, you just put 5 million lives on the books here. What is your strategy for the third year?

Christine Arnold: I do think 2008 gets better for the government from a budget perspective, and a little bit tougher for the health plans, and here is why. That 75 to 80 percent that the government takes over 2-1/2 percent in overage and costs, that declines to 50 percent.

Two things happen in 2008 that right size a little bit the situation that we are talking about with the government, and make things a little bit harder for the health plans. It is going to interject, I think, some rationality into the bidding process. We will see those bids probably rise in 2008.

The first thing we just talked about is the fact that those training wheels, the risk corridors if costs are higher than you expected, start coming off in 2008; 75 to 80 percent goes down to 50 percent. So if you really misbid, government is not going to be there to catch your MLR and your profitability, point number one.

Point number two is that the bids are going to be weighted by the membership, so the companies with experience in this program, not the ones trying to come in with new experiences, are the ones who are going to be very heavily weighted. That was supposed to happen in 2007, and we were all fearful because of the underbidding that is perceived to have happened coming into this year with some plans. Those plans cleaned up on the membership, so when they bid next year, they would have a much higher weighting.

Guess what? They are going to be equal weighted in 2007. A sign of relief, right?5 Things are pretty good. In 2008 they are going to weight it again by the membership within each plan. That means that it is going to be harder, it is going to cost you more because you do not have the training wheels to underbid, and those companies with experience are going to have to use that experience or change the benefits starting in 2008.

I anticipate we will see tightening of formularies. That is the lever that we did not see this year. Ninety-one percent of the top 100 drugs are covered by these health plans, and in the lowest plan, 73 percent of drugs are covered. I think the plans that shot for the membership and cover 99 percent of drugs, they are going to start pulling that lever on the formulary which they have not done yet in 2008 because of these two things that we just factored in.

Robert Laszewski: Christine, just watch out for CMS, though, because CMS has already changed the rules on formularies and it is clear that CMS is more worried about the political outcome in terms of the way seniors are treated than they are health plan profitability.

Christine Arnold: Yes, they are requiring that the home health drugs be covered.

Robert Laszewski: Yes, and they can just keep on doing it. Yes, I think that if you tighten up the formularies you have a better chance to managing costs, but CMS has a track record already of siding on the side of the seniors very clearly in that respect.

Christine Arnold: That would be problematic.

Paul Ginsburg: I think the two of you really put this issue on very nicely about what are the incentives going to be for the plans, and what is the incentive going to be for the government. Do you have anything to throw in on that?

MR. BOSCH: Let me offer a long-term comment, which is this issue of potential budgetary problems in the current drug benefit is also, of course, the issue on the medical side and the nonpharmacy side of Medicare. 7

Ultimately the question becomes what are you going to do both in terms of the prescription drugs and the comprehensive coverage, because right now what we have is an in-between. Medicare, of course, is a compromise between the doctors and hospitals and the government from 1965, and that unmanaged benefit and unmanaged program I think most people agree is not sustainable in light of spending trends.

I think you have two alternatives. Either you turn it over to health plans and insurers and give them some profit margin to manage it through the private sector. Or you look at the other extreme which is going to something that is more like a national health policy that would cover not only Medicare, but the commercial population as well. I don't know that there is a good alternative that is in the middle, and I think the question is broader than just the drug benefit.

Paul Ginsburg: But I wonder if we are headed toward the middle where we will have competition in the private sector, but we will have a lot more aggressive rule setting by the government as to what your formulary can look like.

Matthew Borsch: Right. The potential there is that we go back down the road of the BBA of 1997 where the rule making is in conflict with reasonable business plans from the private sector and you end up with a busted program where the private plans are exiting. I do not see that happening in the next 2 or 3 years, but there is certainly that potential.

Douglas Simpson: I think on the issue of the formulary, the market is going to set constraints around this. It is certainly a level that can be pulled, but to the extent you are signing up people, especially if you are doing an MAPD plan, you need to be very careful.

As to my parents, one parent is diabetic and the other has hypertension and they use a lot of meds. Certainly, if you pick a formulary and one of their meds is excluded, that is not in the interests of the insurer that is underwriting their medical coverage as well, and you can get exclusions, but that is certainly something that is going to come into play.

There is an opportunity there, but I think they are somewhat limited in that ability. I think the bigger thing to watch, it will be interesting to see to the extent that you see very rapid enrollment, we have seen it on PDP this year, Medicare Advantage, people moving from traditional fee for service into private fee for service as kind of a toe in the water and then ultimately Medicare Advantage.

To the extent you beef up that population number substantially, it gives that a critical mass and it will be interesting as approach 2008 to see where politicians are with a large senior population who have embraced these coverages. It is a lot easier to bash these coverages and go after them on the funding side when you have a very small number of people in them. So I think that is the other dynamic that will play out in the political landscape.

Paul Ginsburg: I think you have given me a good transition to start talking about Medicare Advantage. One question I want to pose is, we heard a lot of discussion about the virtues of the integration of the drug coverage with the coverage of hospital and physician services. How real do you think the potential of integration is? Is it just a concept? Or do you think that the plans providing MAPD benefits are really doing something with that integration?

Douglas Simpson: I think ultimately--

Christine Arnold: They are not integrating.

(Laughter.)

Douglas Simpson: I think at this point it has been very much a bifurcated market, but over time I think that is the ultimate end game as to where they would like the beneficiaries to go.

As I said, my parents went on an MAPD plan, I just felt it was simpler for them, and ultimately I think economically they are going to wind up spending up $12,000 a year instead of $17,000. It was a hassle to sign up, to be sure, but ultimately I think they will come out better economically, and then certainly hopefully clinically.

Medicare Advantage, I think the natural progression is to get people involved at the earlier stages of their Medicare life. It is going to be easier to get people aged 65 into the program than it is to convert somebody who is 75 and been in the fee-for-service model for a decade already. Similar to the way we saw 401(k) enrollment in the early 1980s, it was not the 45-year-olds who enrolled in 401(k)s, it was the 23- to 24-year-olds coming out of college, they walked into HR when they joined XYZ Corporation, signed up for their health benefits and signed up for a 401(k).

I think you will see higher takeup rates from the younger population, and over time you will see an evolution in the MA product as those people age into the Medicare population.

Paul Ginsburg: Matt?

Matthew Borsch: I hope I am not jumping on the questions here, but with regard to Medicare Advantage, the really interesting trend has been where the growth has come this year, and it has been not in the traditional local MA-HMO plans where the majority of the enrollment still is, but, rather, in this relatively new private fee-for-service model which according to my understanding the majority of the new MA membership in 2006 has gone into.

The question though, most of it is in rural counties where, if you will, there is a lot low-hanging fruit, seniors who have never had the option of enrolling in an MA plan. In essence the companies are giving them something that is an offer too good to refuse because it has all the flexibility of traditional Medicare plus the coverage of essentially a Medigap policy, but at a substantially cheaper price.

From a straight economic standpoint, it is just a far superior deal to what you can get with Medigap and traditional Medicare combined. Of course, the reason that that is possible is partly because of the artifact right now of these rural counties where you have an arbitrage between the amount that it costs Medicare to provide services to members in the traditional fee-for-service program, versus the national average minimum Medicare payment.

It is not clear ultimately what real value the managed care plans are providing under the private fee-for-service program. It is not clear whether that kind of arbitrage on reimbursement is going to be sustainable. We asked the question to Humana's CEO last week, and they have been one of the biggest growers in private fee for service, "What is the sustainability of this product?" And what he said was we look at this as managed care lite, and so essentially it is a transition product to something that probably looks more like a PPO.

Of course, that then gets into the question of how do you introduce selective contracting and things like that into these rural countries where traditionally that has not worked.

Paul Ginsburg: I'm glad that you answered other questions. I think this is an area that people in this community are not that familiar with, these private fee-for-service plans. Everything you said made sense until you reported what Humana said about a transition into PPOs.

Why would they be bothering? In a sense, they have found a golden goose here. I don't know if any of the panelists have a perspective, is there any value in private fee for service from a broad policy perspective, or is this just a transfer from the Treasury to the Medicare beneficiaries living in these areas with something raked off by these plans?

(Laughter.)

Robert Laszewski: I have been really surprised at the interest on the part of health plans into moving into Medicare fee for service. It is the new frontier. One of my clients is one of the most noble not-for-profit plans out there, and they are headed in that direction very strongly. People see the opportunity to make a lot of money.

I can give you two comments on Medicare Advantage. The first is I am surprised that the growth in Medicare Advantage is as slow as it is. I think we are up by something like isn't it a million lives in Medicare Advantage? I think a lot of that is so much focused on Part B kept people from taking a hard look at it.

I told my relatives who were on Medicare during the Part D exercise that they ought to sign up for Medicare Advantage because it is such a fantastic deal for seniors, and I told them flat out, because the Congress is pouring so much money into this thing. You are getting all these free benefits. You are crazy not to take advantage of it. But they didn't, and it is very much your experience, the leap was just too much.

The strategy that a lot of people are using, a lot of the players in the market are using, is Medicare fee for service because they see it as a really good transitional strategy, and it is a really profitable opportunity.

The second comment that I would make about Medicare Advantage is I think its days are a little bit numbered because MedPAC, as many of you know, came out with some statistics the other day that said in 2005, Medicare Advantage was paid 107 percent of what the standard Medicare plan was paid on a per-member basis, and it is already up to 111 percent. So we are getting a lot of data now, and this is not news to anybody. We knew the Congress put a lot of money into the program to entice people into it. But Medicare Advantage is getting paid a lot more on a per-member basis, and you have a Congress that is going to be looking for opportunities to cut.

If you get a Democratic House of Representatives, it is going to be really interesting in terms of Medicare Advantage and Part D. I do not have to tell you people that the Democrats have both of those programs in their sites. So the data is there that says we are dumping a lot of money into this program, where's the beef? Where in fact are the results that say that we are saving money, and if we are not saving money, why should we be doing it? And politically, you want to cut Medicare Advantage, if you are a Republican, you have to find some ways to trim that Medicare budget.

We have all the pressure on the physicians' side, as you all know. Where is the money come from for the physicians and their fee schedules. So I think a couple years out we have some real problems with Medicare Advantage reimbursement.

Christine Arnold: I think the problem starts next year, actually. There is an overpayment this year. The Medicare Advantage are getting about a 9 percent rate increase in 2006, so this is a great Medicare Advantage year. But part of that is owing to a $2 billion calculation error that they are going to have reverse next year, so it looks like the 2007 Medicare rates are only going up by 2 to 3 percent to correct for the overpayment this year. So I am expecting only single-digit Medicare Advantage enrollment growth next year because they are going to have to cut benefits in order to maintain margins. I am also expect some deterioration in the margin. It is a 1-year hiatus, but that is kind of where we are at near-term.

A couple of observations on Medicare Advantage. Medicare Advantage is the one place I am about adverse selection, because 74 percent of Medicare Advantage have an enhanced benefit relative to PDP, and only 27 percent of enrollment as of early May in stand-alone PDP is enhanced. So if 27 percent of enrollment in stand-alone PDP has the extra drug coverage, and 74 percent of Medicare Advantage has extra drug coverage, then there is potential for adverse selection in Medicare Advantage that I do not worry about in the PDP, given who chose what.

The offset is that over half of the Medicare Advantage plans have zero percent, so 53 percent of the Medicare Advantage plans have no premium. That would appeal to someone who is young and healthy who does not really what the benefits are. That may be the offset, so there is an affordability. But this does not really make sense. We have enhanced the benefits, and yet we are charging no premium. So I guess Bob has a point here that it is probably too good to last.

Matthew Borsch: Could I just jump in with two quick points on what Bob had said?

The first is the 11 percent differential that you referred to, of course, I think a lot of people understand this, but it is important to be looking at this on a comparable benefit package basis I think, because could make the argument from an economist's perspective of longer-term that if under Medicare Advantage you are providing a substantially larger set of benefits and offsetting what would otherwise be seniors' out-of-pocket spending requirements, you are offsetting the need for a certain level of Social Security income. So if I think you are going to talk about the efficiency of Medicare Advantage, you really need to look at it on a comparable benefits basis.

The second thing I would say in reference to the Democrats and what they will want to do with Medicare Advantage, this is a little unclear right now because they have historically been the biggest friends to that program. On the other hand, arguably where Medicare Advantage and Part D generally is making the most inroads and helping the most people is in the base of the low-income and moderate seniors. They are the ones who are benefiting from the government's generosity.

Paul Ginsburg: Let me turn to private insurance for people who are not Medicare beneficiaries. I would like to begin by asking the panel, from the perspective of this audience that we are in front of, what would you say was the most important development in private insurance over the past year, if any?

Douglas Simpson: I guess the first thing I would say is you have to note the consolidation that has taken place in the market, and I think that that has been huge. You have had United take up PacifiCare, certainly WellPoint/WellChoice, we have United John Deere, Aetna has done a bunch of small deals.

So I think the broad theme is consolidation, and we expect that to continue over the next several years. We think about that a little bit more broadly than I think people generally do when they talk about consolidation. To us there are many ways that can happen. It can be large companies buying other large public companies, but beyond that, there is a lot of activity that can take place in the nonpublicly traded market.

There are alliances, joint ventures, partnerships, reinsurance structures. It is going to be very interesting over the next 3 to 5 years we think to watch this dynamic continue. In the commercial world, there are very clear benefits to scale. As the consumer bears more and more of the health care dollar, they are going to demand and place greater emphasis on relative quality of systems and service, in addition to clinical outcomes.

The larger companies have, generally speaking, a better ability to make those investments in customer service and in systems technology. Ten years ago that did not have the same relevance that it has today because people were paying less for health care, but if the monthly premium and co-pays and co-insurance continue to rise as we expect they will, they are going to demand more and be pickier consumers.

There is demand elasticity in health care, and the more out of pocket that you are spending, the more critical of that service that you become. So we think that that dynamic will continue, and I think it will be interesting to watch, and in the Blues universe how some of the smaller Blues wind up dealing with the challenges of consumerism, and I am talking much more broadly than just CDHP. That is one very small piece. I am talking about the general caution of going on to consumers. But I think certainly some of the regions in the Medicare programs did not line up greatly for the state-based Blues plans, and couple that with the encroachment from the larger nationals, it will be interesting to watch and see how that evolves.

Paul Ginsburg: If I can follow-up on the concentration, that presumably this is going to have implications for the people who pay for health insurance and the providers that are paid by health insurance. What is your sense of what we should be looking for?

Douglas Simpson: You bring up a very interesting point. On the provider side, one of the things we hear pretty consistently is that in some of the smaller markets where you once had a dominant Blue with a number of small hospitals, that dynamic has changed.

Now what you may have is a small Blue, UNH and Aetna, and instead of 10 hospitals, you may deal with 3 hospital systems. So the negotiating standpoints have changed, and it is interesting to think about what that means certainly for unit prices. We have heard in some markets that many have commented that hospital consolidation is one of the greater sources of pressure on cost trends for the industry. It will also be interesting to follow this out over the next several years.

I think what is going to wind up happening is the larger companies will continue to gain share at the expense of the smaller companies. This does not happen overnight, but it is a general shift. Each year that leverage works against one-half of the industry and for the other half of the industry. We do not really see that dynamic changing.

Matthew Borsch: I will make a very quick point on the question of further consolidation. Certainly, it seems likely that it will continue to happen particularly amongst the smaller nonpublic plans being bought out by some of the public companies. But I think there is a question as to whether and to what extend the Department of Justice may step in on the next announced public company merger, or perhaps the one after that. We may be a deal or two away from the threshold where antitrust regulators say enough is enough, and it may not be based on just a very narrow market-by-market analysis of concentration, it may be based more on taking a broader view of concentration in the industry. I am speculating, but I think that will be something to watch over the next 12 months.

Christine Arnold: I agree that integration is what is happening. I do not like it. I just get bored. They are all saying the same thing. There has been a total commoditization of the business. The networks have all broadened. Every doctor is in the network because, God forbid the employee benefits manager gets a call from somebody that their doctor is not in network, and then they wonder why the costs are potentially poised to move upward.

I think the benefit of integration in terms of scale is more than outweighed by the detriment of the lack of innovation. When does anything really interesting ever come out of a huge company trying to integrate 30 others? It was the HMOs that came up with something, they were young and scrappy and they were kind of interesting and cool, and those are the companies where things happen that are interesting and that shake things up. That is where innovation comes from. It does not come from -- we have 30 systems here and by next year we are going to have 20. That is not that interesting, and they are all saying the same thing, consumerism. They use different words like health and wealth, or the rise of the consumer, but it is all the same stuff. I think that it is dull. It is.

Matthew Borsch: I think you are right.

Christine Arnold: With employers and consultants, the consultants are like how do I add value walking into a large employer like Morgan Stanley's benefits department when it is Aetna, Signa, the Blues, United? And it has been Aetna, Signa, the Blues, United for the last 7 years. I do not think there is that much of a benefit of scale to the consumer or to the employer, and I think it is time for innovation. I think the market is ripe for some kind of disruption, I just do not know when it how it will happen. I fear that Bob is right and it is the government.

(Laughter.)

Christine Arnold: And then neither the managed care companies or the hospitals will work as stocks, but that creates other issues.

Robert Laszewski: I think Christine is on to a good point, and I absolutely agree with her. I come from the industry, I spent 30 years in the industry, and it has never been so boring or profitable. I guess there is something to that.

I am troubled by all these mergers, especially when you get to WellPoint/Anthem. Those were two giants. There was no economy. There is no scale. What is going on is people who run the big managed care industry are playing to Wall Street, and they have to grow, and they have to grow their profits.

What is so troubling to me is you can grow your profits in a couple of ways. One is you can have all the mergers and get bigger. The other is you can do a better job of managing care, which is what I thought we were supposed to be in. Then you can take a premium for doing such a good job of managing care and controlling costs. I think the industry has given up on managing care and controlling costs, the people who run it, and so if they want to increase their profits by 30 percent next year, they go buy a company that is 30 percent their size, they add it to their portfolio, and now their profitability went up.

Everybody is just as profitable. Everybody has the same economy of scale out there. If you take a company with 1 million lives and you add a company with 300,000 lives, you have just grown your company 30 percent, and that is what Wall Street wants you to do. The analysts love you for it, and you get a billion dollars in stock options for doing that. Why should you do anything else? This really troubles me, though, because I have said this to you before, I think my industry is on a long walk off a short pier.

Where the heck are we in 5 years, or 6, 7, or 8 years. Everybody has the same trend rate. Everybody's costs are going up by exactly the same amount. It is really troubling for me when I look at the longer-term outlook for what we are doing. I think we have a lot of people making themselves rich, and selling our industry down the river.

Paul Ginsburg: We had one really good thought about big-picture consolidation. Are there any others that you want to bring up now as far as a major development over the past year? And realize unlike public policy where you can point to this legislation that passed, this is different.

Robert Laszewski: What I wrote down here to your question, Paul, was continued profitability and little downward pressure on prices I think is the development. Why are we making so much money when we are not delivering any value? I sort of resent this because I ran an insurance company back in the 1980s when you lost money in the indemnity business. These guys are making massive amounts of money in the indemnity business and I am jealous.

(Laughter.)

Douglas Simpson: Paul, one thing I would add is this was an interesting thing, and it is this issue of transparency. We hear so much about consumerism, and you can incentivize people with co-pays and co-insurance benefit design changes, but unless you give the information, what do they do with it, or how do they act? What I thought was interesting is this transparency initiative that Aetna launched in Cincinnati last year where they make information available for common procedures in physicians' offices to individuals, and they can log onto the website and price out different basic procedures.

I thought that was a very interesting real-world application that maybe allows people to become smarter users of health care. It is still amazing to me when I walk around my floor at work and I chat with people and I ask them what they spend on their cable bill, they can tell me to the dollar. And they know if they get HBO it adds $3, if they lose ESPN it saves them $2. They have no idea what they spend on health care, and there is something wrong with that.

Try and call up a hospital and find out what it costs to have a procedure. You cannot. There is just such a huge information lag. So I thought that pilot out of Aetna was interesting. I think we are still a long way away from where people can really be effective consumers of health care. You can still buy a pants much more efficiently than you can purchase health care in this country, and I think that that will persist. But I that was an interesting development.

Matthew Borsch: Could I make a comment both to Bob's and also to Christine's point on innovation? One of the things that has happened with the managed care industry, in a certain sense the providers are, not by choice, but helped create a situation where it is very hard for innovative companies to enter the managed care industry because the time and expense involved with building a competitive network de novo, really nobody does it anymore.

Think about doing it in the New York metro area where you would be trying to replicate something that is as well priced and as broad as what Oxford and Empire Blue Cross have. For all practical purposes, nobody is going to do that. They would probably look instead look to one of the second-tier plans. And I thought it was interesting, I think it was last year, that Aetna paid $400 million for a company in Michigan which was essentially just a collection of provider discounts.

If you think about that, that company, HMS, monetized $400 million in discounts that came from the provider community, and how did the provider community allow that to happen? The negotiation dynamics are such that it is very hard for new players to get a toehold in the industry. Having said that, I would take one issue with Bob's characterization of the earnings picture, to point out that the managed care stocks are down 15 to 20 percent so far this year.

Some of this has been broader health care rotation and the options issue at UNH, but I also think that the market has gotten nervous about what appears to be an uptick in price competition in the industry. So do not rule out competition completely, do not rule out the underwriting cycle, as maybe playing a role, although, arguably, perhaps a consolidation means that margins might come in a little bit but will still remain at relatively high levels compared to industry history. What the market cares about, though, what stocks respond to, is growth in earnings. If you are earning records of $3 billion and it is going to be $3 billion for the next 5 years, that does not work from a stock perspective. Investors want to see it growing 15 to 20 percent every year, and that is some of the pressure that I think you are seeing in the market today.

Robert Laszewski: And so they go buy a company.

Paul Ginsburg: I think this would be a good time to transition to the next topic. What is the outlook for premiums net of buydowns, in a sense if there were no change in buydown, what would premiums be? And do you expect to see turns in the underwriting cycle? Christine?

Christine Arnold: I was cautious on this sector last year, it had performed like it had never performed from a stock perspective, but I thought there were signs of trouble last year.

For the first time last year, we did not have any benefit design changes. Why? Because premium yields fell off a cliff. If the health insurer is going to take 150 basis points off the premium for nothing, why buy down the benefits? So from my perspective, for the first time in 2005, I have a massive divergence between what the companies say they are getting in pricing, and what I am calculating.

Now, of course, I cannot use anything that they report and calculate it myself, I have to rely on what they tell me because this methodology that I have used for the last 7 year is not working. The reality is that pricing fell last year, and it fell 140 basis points from the fourth quarter of 2004 to the first quarter of 2005, 140 basis points.

That is huge. And then it fell another 50 basis points through 2005. So we had 200 basis points, 2 full percentage points, off premium yields in 2005. Unfortunately, while costs decelerated, they did not decelerate that much. So what happened was that the companies, in my view, overreported 2005.

How do we know that? There is this thing called prior period positive development, and it is like talk about boring, but it is a reserve thing. Basically, what happened was these companies had a lot of prior period positive development in 2005, which means 2004 was a whole lot better than they thought it was. They reserved high for 2004 medical claims and they came in lower, and they said, we didn't need that, voom, into earnings, sort of a nontechnical way of reserve development.

So in it came the earnings in 2005, and the prior period positive development came down in 2006, which means that when they went voom in 2004, they did not replace it in the current period, so there is no voom in 2006. So we have an earnings headwind in 2006 because they overreported 2005 because the cycle turned in 2005.

The good news is from my perspective is that we have had a return of benefit design changes in 2006, and we have only seen 100 basis points of premium yield deceleration, and it is actually justified in my mind because we are seeing the benefit design changes resurge. So what I am looking for is continued benefit design changes. If we do not see continued benefit design changes, this sector is in trouble. The good news is we have them for this year.

Matthew Borsch: I would just make one quick comment which is in general I think our forecast for where the premium yield is going to go is probably to be down about 100 basis points, again, excluding benefit changes, resulting from a combination of some continued core deceleration in medical inflation, and also we believe the impact from competition, those two factors together. With respect to where premium yields are right now, an interesting thing to note is that if you look at some of the major publicly managed care companies and you calculate the premium yields from the reported numbers and the reported enrollment, what you get to are lower premium increases than what the companies are talking about.

Christine Arnold: That did not happen in 2004.

Matthew Borsch: That did not happen in 2004.

Christine Arnold: Or 2003, or 2002, or 2001.

Matthew Borsch: And the difference is really between calculated trend yield in the 4 to 5 percent range, versus the companies talking to Wall Street about rate increases of 7 to 8 percent. I think a lot of this difference gets to changes to movement, not only changes in plan design, but movement to high deductible products, different types of products that involve higher cost sharing and lower benefit.

Christine Arnold: I disagree because there are only 3 million people in the whole world in HSAs, and only a million are employee insured. I think it was price competition. It is not benefit changes, and it is not changes in product, in my view.

Robert Laszewski: I come to somewhat the same conclusion, but I look at a different kind of data having been on the operations side. My perspective is we have had high profitability over the last 4 or 5 years because of what I call the trend windfall.

That is, you started out at about a 12 to 13 percent trend rate a few years ago, and in fact, health care inflation has been falling off and it is headed toward about 8 percent, and every year the insurance company is able to charge what the benefits buyer thought was last year's trend. In other words, if the benefits buyer is used to 12 percent and the trend is really about 10 or 11, you can still charge 12, and so there has been this windfall of profit.

Back 3 or 4 years ago there was about a 250 basis point difference between what the insurance companies were charging and what the actual trend was, and that has narrowed now, and it has narrowed to the point where it is probably about zero at the moment. This means that the profit margins are narrowing, and that is starting to show up in some results. But while the profit margins are narrowing, they are narrowing to about what you would expect from excessive, so I think you have sort of a mini price cycle going on right now.

From a policy perspective what I think that means is nothing. You are at unsustainable trends, and for you guys who do policy, whether the insurance company is able to charge 10 percent or 8 percent hardly matters, and that is what it really comes down to. So we are down to probably the lowest trend we are going to see for a few years, which is around 8 percent in reality. The benefits market thinks it ought to be about 8 percent, so there is no windfall margin for the insurance companies. But when you look at what costs are doing, when you look at the physician sector and the hospital sector, it seems to be ticking up at about 8 percent and is where real costs are.

It looks like pharmaceuticals were the reason that we have the lower trend. They came from about 14 to 15, and they are down to about 8, 9, or 10 percent. So we have an 8 to 9 percent baseline trend right now. You may see the insurance companies try to come back and get a little more because Wall Street is a little upset with them. So we are in that 8 to 9 percent trend area for the foreseeable future, when wage and inflation are about 2 to 4 percent.

So we have an unsustainable situation, and it is not as bad as it was a couple of years ago. We have maybe a little bit of a mini underwriting cycle from a profitability standpoint. But the bottom line is we are in that 8 to 9 percent range.

Paul Ginsburg: Doug, what is your perspective on pricing?

Douglas Simpson: We have talked a lot about the pricing side, and we are talking about the relationship between pricing and cost. To switch gears a little bit, one of the things that I think is interesting to think about is what could surprise these companies? What could cause them to wake up and way we really missed this?

If you think about the way in which you underwrite health plans, it uses a multivariate model and you input changes in care delivery patterns, changes in unit prices, and you walk through the new procedures, the new technologies, and you incorporate that into the model along with the claims data that you have. The models do a pretty good job of capturing, and it is a short-talk frequency business, and they are pretty good certainly relative to long-tail casualty coverage. At the end of this model there is a residual error factor which is, among other things, kind of everything else. Part of that are changes in the macroeconomic environment, and that is something we spend a lot of time thinking about because, as I mentioned earlier, as people are picking up more of the tab, the sensitivity of the consumer becomes more important in utilization patterns.

If you look at some of the work done by the RAND group over the last 50 years, they calculated demand elasticity on either price or income of about .2 percent. So every time you move personal income up or down by 1 percent, or you change the price of health care by 1 percent, you effect about a .2 percentage point change in health care change.

It is interesting against the backdrop of this dialogue to think about what has been going on. We have seen very weak consumer disposable income on a real basis. We have seen real wage growth negative, and consumer confidence, unemployment, they are all kind of mixed. The consumer is facing higher short-term interest rates which pressure credit debt and mortgage-related debt, and higher gasoline and oil prices. All these things eat into disposable income and leverage the health care premium.

One thing we continue to monitor is the strength of the U.S. consumer, and to the extent you were to see the consumer get much stronger financially, again, as measured by real wage growth or consumer disposable income, you would see an uptick in health care utilization. There would be a lag effect, but it would come. That I think is something worth watching, and I do not think the underwriting models do as good a job as picking that up as they would like. I think they generally rely on it is a 1-year short tail frequency business and we can pick it up just by renewals.

Paul Ginsburg: If I could summarize some of this discussion, when it comes to the underwriting cycle, there were some discussions about some small cycles, but my observation is how much smaller these cycles are than what we used to have. I think pulling back to the discussion about consolidation, the fact that the insurance industry's entry is much more difficult today than it used to be, probably is a reason that we do not have large swings in the underwriting cycle and probably a reason why on average the industry is more profitable than it historically has been.

I would like to go to the next topic which is developments in network design and provider payment strategies, and I really want to bring up two topics. One is high-performance networks, and the other is pay for performance for hospitals or physicians. Since I read your reports and I read disclaimers, there is a disclaimer that HSC is involved in a study of high-performance networks funded by the California HealthCare Foundation. Otherwise, what is your perspective on how important the developments as far as potential high-performance networks is?

Matthew Borsch: I would quickly say the jury is very much out on this. It has not been established yet. The potential is there, but it requires employers and health plans to have real backbone in making narrow networks within networks work by providing very strong incentives for members to use them. We have not seen any signs that the health benefits community is ready to really push that in a strong way that would lead to widespread effectiveness and adoption.

Paul Ginsburg: Christine?

Christine Arnold: According to the Mercer data, about 22 percent of large employers have these kind of high-performance networks, and if you talk to the benefits leaders, some of the consultants, it is definitely top of mind. That and health advocacy are the big things. The problem is, and I am not sure what health advocacy means, but I am going to give you my definition of what health advocacy means because it has yet to be defined.

I do not think having a high-performance network, and I am sure we have them at Morgan Stanley, like if I got diagnosed with something, would that be the thing I was looking for? Probably not. So I can go to my Mayo website and fill out this health assessment where I say I weigh 120 pounds and then they do not call me to tell me to lose weight.

(Laughter.)

Christine Arnold: You can game them. That is not that funny.

(Laughter.)

Christine Arnold: So I have the Mayo thing, and I am sure you have done that. They ask you if use seat belts and that kind of thing. Then I have the nurse help line that I can call if my kid has a fever or has spots all over and it is on a weekend and I do not know what to do. Then I have my health plan that I can call. Then this is just another thing to add to the last. Great.

Now we have special networks, so if something happens to me, there is a neonatal network, there is a cardiovascular network, I do not even know which ones I have. There has to be some kind of integration. So this is where the one thing that we are talking about, which is health advocacy, could play a role if properly executed.

Shouldn't there be one number that you can call where somebody who has some medical knowledge, a nurse, a nurse practitioner, a whatever, can integrate all this? I think for this to have any legs at all, we need to stop adding new things, and this is a new thing, and start integrating all the things. And disease management, too, right? There is disease management out there. I have just named five or six things that I know we have at Morgan Stanley which are completely unintegrated, many of which require a separate password if I get onto our intranet, and I cannot even find them.

Shouldn't I be able to call one phone number and that person pulls up my benefits and says, Christine Arnold, you have just been diagnosed with XYZ. It is really unfortunate. But you know what? We have this high-performance network, and let me education you about what that means, and let me help you. Maybe you should get a second opinion. Maybe getting a second opinion is going to take too long.

Somebody needs to help. Right? So I think the high-performance networks together with some way to integrate all this stuff has tremendous potential in improving people's health. There are also studies out. Arnie Millstein is all about 40 percent cost savings, right place, right time, high performance definitely has potential, but it is daunting being sick.

Robert Laszewski: I would add on pay-for-performance networks, one of the major Northeastern Blues plans, and this is typical, I will not say which one, say they are going to spend $189 million this year on P for P.

Paul Ginsburg: You are talking about P for P, and not high performance?

Robert Laszewski: I am going to make the same point, P for P, and pay for performance, and they are going to measure how well doctors adopt programs to reduce errors, grade hospitals in patient satisfaction surveys, and they are going to grade hospitals on 5 criteria that the hospitals picks from a list of 14. And how would you like to be graded if you get to pick the 5 out of 14?

The point is, and this gets back to high-performance networks or pay for performance, it comes down to exactly the same thing, and that is we have no mutually agreed-upon criteria between the payers and the providers on what matters. There is no data that really drives you in the right direction.

In high-performance networks you say this is a high-performance doctor or hospital, but you do not really have the data to back that up, it is almost more intuitive. There is just simply no agreement on how you reduce costs. At the same time, you saw I am sure at Dartmouth Medical School, Jack Wennberg has been doing this data for 20 years now that I know of.

Paul Ginsburg: How about 40?

Robert Laszewski: How about 40, but 20 that he has been publishing good stuff. He says, among academic medical centers, in the final 2 years of life care costs vary by a factor of 2, the admit rate by a factor of 5, and the average days varies by a factor of 3. So we have this wide variation, and somebody spending $189 million on patient satisfaction surveys and letting hospitals pick 5 criteria out of 14. This is all fluff.

Why should anybody get serious when you can make the money you are making not rocking the boat? And benefits managers are willing to pay it. For some ungodly reason at $11,000 a year average family cost of health insurance, it is not enough to get people serious. Pay for performance and the high-performance networks are going in the right direction, but there is no traction.

Paul Ginsburg: If I could interrupt you, we are going to go to questions soon, so if you could pass any question cards you have to the aisles and the staff will pick them up, and we will prepare for the Q and A session.

Douglas Simpson: So I do not get the hook here, I will be quick. The one area that is kind of interesting and I heard all the comments that were made previously, but it is kind of interesting on areas like specialty pharmacy, areas that do not lend themselves to underwriting fixes. You cannot solve a $60,000 script with co-insurance. To do it in a meaningful say, you are effectively not insuring it for a large portion of the U.S. population. Slap a 30 percent co-pay on a $60,000 drug and it is $18,000, and you cannot do that.

Some of those types of issues, it is kind of interesting to think, is there a way to structure a network and use steerage to help address those types of issues. But I also keep in the back of my mind that this is an industry where I get a phone call saying that the reason my son was not covered is that he opted out of our health plan, which is pretty amazing because he was 3 months old at the time.

(Laughter.)

Douglas Simpson: So while I was certainly proud as his father, I was a little surprised. I guess he was protesting the premium increases.

Paul Ginsburg: Are there any other thoughts on network? Is there anything you want to say on consumer-driven health care in general while we are waiting? And people can start coming up to the mikes now.

Matthew Borsch: If I could just offer a quick observation on consumer-driven health plans, and it is really a global point. The big question is whether or not the employer-sponsored system of coverage is going to unravel over the next 5 to 10 years, and there are at least two sets of pressures that say that it might.

You have small employers continuing to either not offer benefits, offer partial benefits, or drop coverage. You have more employers pushing enrollment into their self-insured plans or become self-insured, and opting out of the broader underwriting risk pool. And now you have consumer-directed health plans that have the potential to segment and unravel the broader insurance pool that small- and middle-sized employers rely on even further.

I do not have the answer to that, but I think that is going to be a very big challenge over the next 5 years.

Robert Laszewski: Paul, I would say that I think we can get down on the fact that it is getting worse on all fronts and we have not accomplished anything from a public policy perspective. I would just like to say, though, I think there are three important experiments going on in the system right now. One is HSA/HRAs, consumer-driven care, the other is the Part D drug benefit, and the third is the Massachusetts bill.

Those three things I think are very, very important, because we are going to have a serious public debate on health care sometime in the future, and the results will be in on those three things. With consumer driven, obviously, the Republicans and the free-market forces, consumer choice, moving the individual purchase of insurance.

Part D is very important as an experiment because there is the debate about how you bring Medicare costs under control and whether privatizing Medicare is the way to go, and you have privatized a big Medicare benefit and the results will be in 3 years and we will know whether privatizing Medicare makes sense based on what happens to Part D.

The third is Massachusetts because you have a universal mandate. I hope Massachusetts gets implemented. My concern is that it may not because I think the Achilles' heel is cost containment and being able to come up with an affordable plan, but you have an experiment about all the things that people on the other side of the spectrum believe in. So there are three very important experiments going on that are going to give us some answers when we do in fact get serious about this.

Paul Ginsburg: Let me turn to some audience questions. One actually that was set up by these two comments, Is employer-sponsored health insurance sustainable, or will employers stop offering it to their employees? Does anyone want to take a crack at that? Let's try to do one answer to each question so we can cover more.

Christine Arnold: Seventeen percent of small employers entering this year increased the premium to the employees, the portion of the premium paid. Keep in mind that only 60 percent of employers that are in the small-group market offer health insurance at all, and on average they require that more than half the premium be paid by the employee.

So we have seen a movement away from employer-sponsored coverage in the small-group market. We are not seeing that in large-group. And actually, the deceleration in premium yields and benefit trends I think pushes that off. We are seeing calculated cost trends in the 5 to 6 percent range, so while not at the level of overall inflation, we are not at a breaking point right now.

The reason larger employers are going to self-insuring from fully-insuring is because they believe that the trend is decelerating, otherwise, if they thought it was going to uptick, they would not take that risk on their income statements. So we are not at a breaking point yet. If trends reaccelerate, which is the big risk to managed care earnings because they will not anticipate it, it will just happen. Or if we see premium yields turning up, which probably will not happen on its own. This is something a year or two out. That will create, I think, more of an impetus for this, but right now things are actually okay.

Paul Ginsburg: I have a question about association health plans. What is your sense if legislation like that should pass as to what implications would that have for the industry?

Matthew Borsch: Just to answer very quickly, I think it would be another step along those lists of things that would segment the underwriting risk pool if not only that legislation passed, but there was a lot of takeup of association health plans amongst small employer groups. You would have that on top of CDHPs, on top of the self-insuring trend amongst the larger employers, to segmented, and ultimately I think undermine the underwriting risk pool.

Paul Ginsburg: Bob has spoken a lot on this issue.

Robert Laszewski: Yes, I am a repentant cherry-picker from the old days.

(Laughter.)

Robert Laszewski: People talk about passing an association health plan bill, and I think as many of you know, there is no law against association health plans. I ran a number of association health plans in the 1980s and the early-1990s, and you could start one tomorrow.

The demise of the association health plan business that I was in was HIPAA and all the state underwriting changes. That is what put the association health plans out of business. There are a few around, but not very many. It would just screw the market up royally. It is about segmentation, it is about back-door segmentation, and all the things that the people who are against it have told you about segmentation is absolutely right.

What was ironic in this debate is the way the Blue Cross plans pulled out of the opposition, and the industry was segmented into two parts. The big plans were willing to see the AHP bill pass because the Senate version gave them the ability to do it. And the smaller or the regional plans that were the one-state Blue Cross plans were scared to death because the big national plans knew they could come into Mississippi and cherry-pick the heck out of the Mississippi Blue plan. All I can tell you is I ran those things years ago, and they are a cherry-picker's delight.

Paul Ginsburg: We have another question. Why did capitated payment systems for physician and hospital organizations fail?

Christine Arnold: They failed because they separated the process of underwriting with the consequences of underwriting. So if I can set premium yields down 10 percent in order to grow membership but you have to live with the 10 percent payment cut, then I can grow the membership, maintain my margin, and you the doctor or hospital have to live on less.

The problem became most acute in the last down cycle. In 1997 we had a lot of capitation in Medicare and payments did not keep up with cost trends, and in 1997, 1998, and 1999, capitation fell apart particularly in the Medicare segment, but in the commercial segment as rates did not keep up with cost trends.

Paul Ginsburg: Given consolidation in the managed care sector and premium yield declines, how have providers continued to garner such strong pricing increases from managed care payers, and when does that change?

Douglas Simpson: I think one of the dynamics we touched on earlier was certainly in the hospital community provider consolidation. We have heard about that now for years, and that will likely continue. We have seen a change in the negotiating leverage of the providers in many markets with the advent of increased competition among the payers and increased consolidation among the providers. There are certain markets in Pennsylvania where I know point blank there was an individual payer negotiating with multiple hospitals, and that dynamic has just 180 degrees flip-flipped, and it has real implications for unit cost trends.

Robert Laszewski: What created the indemnity market we have today was the patient rates per billion. In the late-1990s when you had the providers and the patients rebel and the benefits managers did not back up the HMO industry and the HMO industry learned it could make more money just passing increases through, we went to the indemnity model.

We moved away from the managed care model and we moved to the indemnity model. So we received a truce between providers and health plans. The benefits managers are willing to pay for it, and we have had enormous profitability. The providers are doing better, the plans are doing better, the benefits managers are willing to pay for it, and something has to change there before we get serious again about managing costs.

Paul Ginsburg: I have a question about the impact of obesity on the industry.

Christine Arnold: The impact is that we all lie on the Mayo survey.

(Laughter.)

Christine Arnold: There has been a move afoot among some of the larger employers like GE, even Smith Barney and Xerox, to start to award points for changing behavior and lifestyle. If you smoke and you complete a smoking cessation program, get a coach and get tested and you do not have any nicotine in your blood, great. Or if you get your BMI down to a certain point. Or if you go to an NCQU accredited physician as your primary care doc or some other criteria, employers are starting to award points which reduces either the co-pays or deductibles that you have to pay, or the premiums that come out of your paycheck. We have also, of course, seen things like $25 for each completion of a health risk assessment that then goes to the health plan, and they can attempt to enroll you in the disease management program.

Paul Ginsburg: I have a good question, and it is a good final question. Why can't cost trends go much lower? It was close to negative in 1996.

Douglas Simpson: One thing is I think it could go lower, stop innovating, and I think that is a big part of this. I think when people talk about medical costs inflation, they equate it to CPI, and I think the reality is there is quite a bit of a difference.

One of the not-for-profit health plans we deal with was speaking with an audience and they said, Why is the inflation so high? Can't we get it down? And he said, yes, we can give you surgery and we'll take out your gallbladder and you will have a 6 inch scar. You wanted laparoscopic surgery.

You could do those types of things, and nobody wants to. So you buy a car this year and you buy the same car next year, it does up 3 percent, that is inflation. If you buy the same car next year and it has a DVD player, air conditioning and fancy tires and rims, it is going to cost more than the 3 percent increase. I think that is the one thing you do not see mentioned very much in all the discussions about medical cost trends is the extent to which there have been improvements in care, and that is a big driver of this.

It is interesting what Americans are willing to pay for and what they complain about, but there has been substantial innovation and it comes at a cost.

Matthew Borsch: Just a quick point on that. Looking back at the cost trend low in the mid- to late-1990s, you have to factor into that the reflection of the benefit of the large shift to managed care and one-time trend benefits that came with that wholesale shift into managed care. You could see that happen potentially again with a wholesale shift into consumer-directed health plan products.

We really do not know what the growth trajectory is, if in fact we are going to move wholesale to CDHPs, if it is going to be on a 5- year or 15-year time horizon, but that would clearly be something that would have the potential to do it. Whether the consequences in terms of all of the out-of-pocket spending would be acceptable or not remains to be seen.

Paul Ginsburg: I think at that point I would like to thank this panel for a phenomenal job, and we will restart at 10:45.

(Recess.)

SECOND PANEL

Paul Ginsburg: I would like people to begin taking their seats. We are ready for the session on providers, hospitals, physicians, and pharmaceutical companies. I probably ought to begin by asking about the first question. The first set of questions is about underlying health spending trends. I am talking about what has been going on with hospital spending and what is your outlook for the near future, and first ask a general about what is the outlook for hospital spending and what are the drivers of it. Chris, do you want to start?

Christopher McFadden: Thank you, Paul. It is a pleasure to be back again this year. I would say starting with the hospitals is to have a more modest kickoff to the panel.

Certainly, unit volumes in the hospital industry continue to be very modest, driven by a variety of factors. Driven by the ongoing migration to outpatient settings, driven by what continues to be higher levels or certainly high levels of patient sensitivity to costs, and that is having clearly an effect on admission patterns. And I would say a third factor that has come to the surface in the last 6 months or so is the perception at least that perhaps physicians are modify some of their practice behavior in a way that defers hospital admissions within an electable course of treatment strategy both because of the mechanics of trying to practice in an office-based setting and in a hospital-based setting, and perhaps some lost efficiencies that come with those tradeoffs, as well as sensitivity around the malpractice environment where malpractice is again a topic at least in the physician community that has been as sharp concern and perhaps there is some perception that there is more risk associated with an inpatient than outpatient experience.

I think it is hard within the alchemy of those issues to know what is the most important factor, but at least those three factors are having an impact. What is interesting, when you look at the data is actually revenue per admission is increasing, and that is not unintuitive, because if what is happening is a cohort of slightly sicker patients or more developed cases are what is being represented in the admissions pattern, then the revenue pre combined with some of the pricing increases which have consistently been in the mid to high single digits will drive higher revenue per admission even if admission trends in the absolute sense remain relatively modest.

John Wells: I think in the hospitals that I cover which, as Paul mentioned, is primarily the not-for-profit world, we are seeing cost trends and spending trends accelerate at a fairly healthy pace. I think the drivers of that continue to be pension expense and largely bad debt expense, but also as we look at on-call coverage for many hospitals, hospitals having to pay for physicians to provide coverage in the ED is certainly impacting spending trends, as well as capital costs and capital expenditures.

Maybe to a lesser extent we still see supplies, labor, and insurance as being significant cost drivers. Labor is not as concerning as has been in some of the previous years, but we do note that as a percentage of revenue, labor and benefits expense has not fallen below 51 percent of hospital revenue since 2000. Then on the physician side, primarily we are seeing physicians still struggling with liability insurance and certainly getting squeezed more on the revenue side.

Paul Ginsburg: One thing that struck me is for years people have commented about the hospital building boom, about how many cranes are up there, how much construction is going on. But it seems there is a disconnect between this discussion of spending trends and all this capacity that is being built. Are there any comments on that?

Geoffrey Harris: I would add to the previous question, and then I will address this one. Admissions trends have been surprisingly weak at least relative to expectations over the last couple of years. Often admissions trends tend to lag economic growth by a couple of years, but for some reason, and I would like to list a few, the admissions trends have really not picked up even a couple of years now after a rebound in economic growth from the slowdown in the early part of this decade.

I think there are probably five things, a couple of which have been mentioned, but maybe five things that are affecting this. One is changes in benefit designs, shifting more costs over to consumers. Secondly, rising uninsured. Third, and probably very importantly, competition from other delivery sites, namely, outpatient facilities, surgery centers and that sort of thing. Fifth, I think some of the disease management initiatives that were somewhat panned in the previous panel are having some impact in terms of hospital utilization trends and may be shifting to other settings. Then fourth, and again somewhat surprisingly, I think new technologies are typically associated with increased medical utilization, but I can think of at least one instance, for example, drug-eluting stents, where they have actually resulted in a reduction in certain cardiac procedures due to lower restenosis rates, in other words, patients who would otherwise come back for another procedure are not anymore.

So I think there are a series of factors that have influenced admissions trends. I think in that context it is surprising that there has been a hospital building boom, although I think a lot of that has been focused in outpatient capacity as hospitals have tried to address changing market demand.

Christopher McFadden: And I might just build on Geoff's comments by saying relative to the construction environment, when you look at some of the data, and the Census Department of all entities has put some interesting data about this trend has looked like over the last couple of decades, one is, you have a certain replacement cycle. When you think about the stock of hospitals in this country that were boluses of building over time, there certainly was a bolus of building 35 to 50 years ago, and so you have some upgrade and replacement of that bricks-and-mortar infrastructure.

But I think the other element, and Paul I know will touch on this a second in the context of DRG, refinement proposals that have come out of CMS, if you think over the last several years what has been really important to the average hospital CEO has been the admissions pace, particularly in some high-income categories like cardiac and orthopedic, spinal and some other key categories.

One of the ways that you compete for physicians in those areas, particularly in more densely populated urban markets where they have multiple admission privileges is to make sure that your hospital has the latest and greatest stuff, and that may be the latest and greatest imaging technology, that may be the latest and greatest surgical suites, and that, too, I believe has contributed to some of the capital spending that we have seen.

It will be interesting and it will hard to prove except for a couple of years from now whether this realignment of some of the reimbursement methodologies under the DRG program perhaps shifts that incentive modestly as the income potential, at least within the cardiac area, is not what it was historically.

Robert Berenson: I would just pick up on those comments and refer to the work from the Center for Health System Change, to give you a little plug, has done this past cycle.

I was involved with interviewing a lot of hospital executives, and we are going to have an article published next month, about the increasing service line strategy that hospitals are adopting where floors are no longer med surg floors, they are heart centers, oncology centers. In one place with an educated population they are not advertising genital-urinary problems, come here, they have branded it a pelvic floor dysfunction center, a sleep center.

(Laughter.)

Robert Berenson: So there is sort of an appeal to consumerism, not the part of consumerism that expects people to use their money to theoretically reduce costs, but the part of consumerism that wants the best and the greatest. And I agree completely with the disability of attracting physicians to the newest and the best as well. We are characterizing this as a continuation and an accentuation of sort of a medical arm's race where the competition is not overpriced, but it is over essentially the state of the art and branding that state of the art in a way that consumers and physicians will respond to. But the flip side of that, and, again it picks up one of the comments made, is that physicians are getting into the same business.

These would be examples of disruptive technologies which permit a lot of things that used to happen inpatient or even outpatient, but within the hospital environment not can safely be done in ambulatory surgery centers, or increasingly, even in physicians' offices, and physicians in many parts of the country are getting very entrepreneurial.

So you have simultaneously hospitals trying to develop close relationships with those physicians who they need to market a service line, and in some places that means actually employing and giving incentive bonuses for physicians to part of that service line strategy, and other physicians trying to compete with that service line by doing it on an outpatient basis outside of the hospital orbit with their own facilities.

We can talk a little bit more about that, but I think that is an interesting dynamic when we can talk about whether that whether that will promote price competition or not.

Paul Ginsburg: I a few months ago published a paper which was focusing on future demand for hospital services, and the purpose of the paper was to say that aging is not as large a factor as people think. In subsequent interviews with the media, I started saying, and I want to run this by the people on the panel to see if I was right, the biggest uncertainty in hospital capacity planning for the future is how much of the outpatient work you can expect the hospital garner and how much of it is at risk of being lost to physician-owned ventures. Do you have any reactions to that?

Geoffrey Harris: The main comment I would make is that as long as physicians are allowed under Stark legislations to set up and own physician-owned entities, I think that that is a trend that will continue and be a significant challenge to hospitals.

I can think of one example in a particular market in Texas that happens to be represented by a number of for-profit companies where an entrepreneurial physician group set up a specialty hospital as well as some additional facilities and has taken all of or a significant percentage of the profitable patients from the general hospitals in the arena. So they are suffering, and the physician-owned hospital is full. As long as that is allowed legislatively, I think that will be a continuing challenge.

Robert Berenson: Just to pick up the regulatory environment on this, one is certificate of need, and, obviously in a lot of states there is not much certificate of need left.

It was interesting to talk to health plan executive who in the context of specialty hospitals became advocates of states adopting certificate of need because they did not actually see the competition that was being created by specialty hospitals as an alternative to community general hospitals as leading to price competition.

There was some disagreement amongst the plan executives we did interview, but at least a significant number of them said we want certificate of need, and some states have actually gone that route. The second piece of this is Stark, and it is real clear that there are lots of exemptions. The clearest one is that Stark does not apply to a physician practices which purchases or leases the equipment itself, and it is not just referring to another facility where the physician may have some ownership.

So one of the trends that is going on is single specialty physician mergers not primarily anymore for the sole purpose of having more leverage and negotiating rates with managed care plans which has been a strategy that single specialty groups have done, but, frankly, to have scale to be able to refer enough patients so that an oncology group might purchase a PET scan or lease a PET scan, or a neurology-neurosurgery group might do the same thing with an MRI machine. So there is a merger so that there is enough volume to self-refer, and there does not seem to be a specific regulatory barrier to that.

I think most people are looking at correcting some of the pricing distortions at least on the Medicare side that make certain services uniquely profitable and others not so profitable, and that might be one approach.

Christopher McFadden: If I could, just one additional comment because I think that the term entrepreneurial instincts or incentives on the part of physicians are obviously an integral part of seeing ambulatory strategies or outpatient strategies continue to migrate and be successful.

One of the things that is interesting to think about is the complexion of medical students in residency programs, because what one sees in those programs is a fairly meaningful demographic shift, a much, much higher percentage of women are in those programs either as part of dual professional families or dual physician families.

And there has been a lot of discussion that the next generation of physicians may be more inclined to want to be employed and 9:00 to 5:00 or 9:00 to 7:00, as opposed to be trying to run these expanded enterprises of their inpatient business, their practiced-based business and their ambulatory surgery center, minority interest on the west side of town and their minority interest on the east side of town, because that takes a lot of effort and a lot of energy. It does not say that people do not have energy and effort, but it might say they want to apply it in a different way or achieve a different work-life balance.

So I think as we think about the combination of that element, the reimbursement element, and the costs of doing business element, I think you could see a dynamic in which you get a little bit of a polarization. You see some of that outpatient or ambulatory work flow back into hospitals, and as technology continues to enable, perhaps more flow into the pure practice, and that middle tranche, what we think about as things like ASEs and dialysis and imaging centers, may be a slow-growth part of this continuum of care that we see in the marketplace.

Paul Ginsburg: That is an interesting point, and maybe if I could ask Bob what percentage of physicians need to be entrepreneurial as individuals to get into a very entrepreneurial physician sector?

Robert Berenson: I do not have that answer. What percentage of your practice needs to be capitation before you flip to practicing differently, that is that kind of question and I am not sure I know the answer. But I was going to make a comment on this one, at least some physicians are becoming entrepreneurial, and there are always come in a community who have been venture capital of various kinds that is out there, and, again most of the attention is on specialty hospitals, but the bigger thing is on other things, imaging centers, ASE's, other kinds of activities like that.

The hospital then reaches out sometimes not happily, but often half the loaf is better than no loaf, so the hospital reaches out and said we can do a joint venture with you. We will bring the managerial stuff to the table, we will bring our capital to the table, and so physicians who do not want to be entrepreneurial may be in, who knows if this is good or bad for the world? Is it a focus factory?

Some hospital executives actually say it has been good for what they are doing that they have this potential competition from physician-owned ventures, that they have woken up to the need for greater efficiency and responsiveness, both to doctors and patients. Is it a focus factory or is it just a branding tool to get more patients is the open question. But at least in some communities, you actually have the hospitals and the docs joint-venturing and you do not need quite the same entrepreneurial doc out there. But you do need at least some sort of entrepreneurialism in the physician community, but I do not think it has to be a dominant factor.

Paul Ginsburg: I was just thinking of a physician that I know that is entrepreneurial and runs a ten-physician group, and the other nine physicians do not have to be. They can jus