明晟 (MSCI) 2009 Q3 法說會逐字稿

Good day and welcome to the MSCI third quarter 2009 earnings conference. Today's call is being recorded. At this time, I'd like to turn the conference over to Mr. Michael Neborak. Please go ahead.

Thank you, Operator. Good morning and thank you for joining our third quarter 2009 earnings call. Please note that earlier this morning we issued a press release describing our results for the third quarter 2009. A copy of that release can be viewed on the Company's website at MSCIbarra.com under Investor Relations.

This presentation may contain forward looking statements. You are cautioned not to place undue reliance on forward-looking statements which speak only as of the date on which they are made which reflects management's current estimates, projections, expectations, or beliefs, and which are subject to risks and uncertainties that may cause actual results to differ materially. For a discussion of additional risks and uncertainties that may affect the future results of the Company, please see the description of risk factors in forward-looking statements in our Form 10-K for our fiscal year ending November 30, 2008.

Today's earnings call may also include discussion of certain non-GAAP financial measures. Please refer to today's earnings release for the required reconciliation of non-GAAP financial measures to the most directly comparable GAAP-measures and other related disclosures.

Since we will be referring to run rate frequently in our discussion this morning, please let me remind you that our run rate at a given point in time represents the forward-looking fees for subscriptions and product licenses that we will record over the next 12 months, assuming no cancellations, new sales, or changes in the assets and ETFs licensed to our indices. Please refer to table 2 in our press release for a detailed explanation.

Henry will begin the discussion with an overview of the third quarter, then I will provide some details on our financial performance. Following our formal remarks, we will open up the line for questions. Our prepared remarks will focus on commentary that is incremental to the earnings release so as to leave adequate time for questions.

I will now turn the call over to Henry.

Thanks, Mike, and good morning, everyone. As you saw in our press release, we reported Q3 revenues of approximately $109 million, record adjusted EBITDA of $54 million, and a record adjusted EBITDA margin of nearly 50%. Excluding a one-time foreign exchange adjustment, revenues were $112 million for the quarter. These results highlight the high level of profitability inherent in our business as well as the diversity of our revenue base across the globe. Despite our revenue growth continuing to be sluggish.

Our equity index asset-based fee business had an excellent quarter as a result of the continued rally in a number of equity markets around the world and continued inflows into exchange credit funds. These trends have continued into Q4 and the assets under management balance in ETFs linked to our indices closed September last night at a near record level of $215 billion.

One of the great aspects about our business is the complementary nature of our asset-based fees and our subscription businesses in operating environments such as the current one. Our subscription business tends to lag the market environment while our ETF fee business is more of a leading indicator for the operating environment. This was evident in our Q3 performance as we experienced strong growth in our ETF fees when our subscription business was flattish on a run rate basis. In Q4 of last year, and in Q1 of the current year, the reverse was true with our ETF fees under significant pressure and our subscription business still growing.

Within the subscription business, we continued to see a divergence in trends among two sets of clients and between our new sales activity and our retention rate during the quarter. On one hand, there is a set of clients around the world that have moved beyond the crisis and are trying to expand their business, and consequently, increasing their budgets and buying new tools from us. This is evident in the relative strength in our new sales which reflect the areas of strength throughout the investment industry, particularly (inaudible) funds, whose budgets appear to be less sensitive to the decline in asset values, and in opportunistic small and medium sized asset managers around the world.

On the other hand, there is another set of clients which is still grappling with the repercussions on their business from the financial crisis, and are still cutting costs and restructuring and maintaining tight budgets, and consequently, reducing or canceling certain products from us. The retention rate weakness that we have reflects the continued focus by this set of clients, particularly large asset managers, on managing their expenses very tightly and in centralized ways.

An important positive factor in our subscription business during the quarter remains our new sales activity, as a result of a high level of focus from us on clients that have budgets to spend and determine efforts that we have to bring products to them. New sales increased sequentially for the second straight quarter. In Q3 we generated $16 million of new recurring sales, which is up from $14 million in Q2, and $11 million in Q1. The increase in new recurrent sales was led by our equity analytics and multi-asset class analytics product categories, reflecting continued demand for our portfolio management and risk management tools.

The retention rate for Q3 declined to 81% from 92% in Q3 '08. Approximately 40% of the $17 million of cancels that we recorded in the quarter were due to budget constraints. Another 20% was from a change in strategy or organizational structure by our clients, such as the elimination of a proprietary trade index. Nearly 15% was due to merger activity, and approximately 10% was due to price reductions resulting from contract renewals. Importantly, less than 10% of the cancels recorded during the quarter were for customer dissatisfaction with our products. The balance in the retention or cancels was largely due to product swaps. The weakness overall was most pronounced in equity analytics, which I will elaborate more later on in my remarks.

A month into Q4 we see signs of easing budget constraints among some of our subscription clients. Sales activity remains healthy particularly for BarraOne and the quality of the pipeline for all of our product has improved from Q3. However, in Q4, we will likely continue to experience the pressure on our retention rates that we saw in Q3 as the operating environment for some of our clients remains challenging and the level of cancellations remains difficult to predict. For equity analytics, which, as I said, had a particularly challenging Q3, the retention rate for Q4, based on currently available data, looks better than Q3. But it is still early in the quarter and premature to speculate on how the full quarter will ultimately turn out.

As you saw in this morning's release, our adjusted EBITDA margin reached a record 49.6% and again exceeded our long-term financial target of the low 40s. The reasons for this outperformance are similar to what we discussed in our Q2 call. Mainly, strong growth in ETF fees, disciplined and focused expense management, the high margins inherent in many of our products, and favorable benefit from foreign exchange rates, and a keen focus on growing the absolute level of adjusted EBITDA rather than maintaining a target margin in a period when revenues are increasing at a rate below our long-term growth target of mid teens. As long as these factors that I just described remain in place, and as long as ETF assets remain at or above current levels, it is likely that our EBITDA margin will exceed our target. However, keep in mind that we are continuing to hire new employees around the world, which will increase our compensation expense in the balance of this quarter, and throughout 2010.

Now let me update you on our investment spending plan. As we mentioned in our call last quarter, we had approved the hiring of approximately 100 people. Since that time, we have filled about half of these positions. In addition, we recently approved the hiring of another 65 people in the Company. Largely in our emerging market centers and primarily in the areas of sales and client support, product management, research, information technology, and data operations. This investment will enable us to continue to scale up our client coverage worldwide, enhance our existing products create new ones in order to accelerate revenue and profit growth and deliver value to our clients and shareholders.

Another highlight in the quarter is our free cash flow generation. Our net debt position at the end of Q3 was $23 million, which is down from $73 million at the end of Q2.

Let me now go deeper in terms of our different product categories. In my prepared remarks, I will be referring only to run rate figures, whereas Mike will focus on accounting numbers. I will be speaking to comparisons to Q2 of this year unless otherwise noted. Run rates are only a snapshot of our business at a point in time and only serve as an indicator for future revenue.

In terms of major product categories, the run rate in equity indices overall increased 6.5% compared to Q2. Equity index data subscriptions recorded a 2% sequential increase in the run rate. Growth in our core index modules increased to 2% compared to 1% growth in Q2 and relatively flat performance in Q1. We saw solid growth in emerging markets, small cap, and the developed market modules. We also saw acceleration in custom indices and user fees, which were up 8% and 2%, respectively. We achieved 7% growth in reporting fees. The benefit from price increases enacted some time ago contributed about half of the sequential run rate growth.

We had our best quarter in terms of new client activity as measured by sales and client additions in equity and data subscriptions since Q3 '08. Sequential growth in the run rate was seen across all regions with particular strength in the Americas and all client segments with the exception of hedge funds. The aggregate retention rate for equity index data products declined to 91% from 96% in Q3 '08. Approximately one-third of the cancels reflect structural changes among our clients, such as mergers, or closures of a firm or a fund or an office. Just under 25% reflect budget pressures, and 20% reflect price reductions. Less than 10% of the cancels were due to customer dissatisfaction.

The other part of our equity index business, our asset-based fee business, I will defer to Mike to provide further color on this in his remarks. The run rate for equity portfolio analytics declined 4.3% sequentially. The divergence, or tug-of-war, between clients who are expanding and those that are still restructuring is most dramatic in our equity analytics business. New sales activity for equity analytics was encouraging and continues to increase. New sales were higher in Q3 than Q2, as we saw some easing of budget constraints as well as some pent up demand translating into new sales. New sales in Q3 were $4.5 million in equity analytics, compared to $3.9 million in Q2, and $2.7 million in Q1.

In addition, our efforts to upsell existing clients yielded benefits in the quarter. In addition, we added 19 new clients on a gross basis, which is the best quarterly performance since the third quarter of 2007. Unfortunately, cancellations in equity analytics continue and accelerated in the quarter to $10.5 million compared to $5.8 million in Q2 and $4.5 million in Q1. The aggregate retention rate declined to 68% from 88% in Q3 '08. The weakness was most pronounced in the Americas and was broadbased across asset managers, broker dealers, and hedge funds. Weakness in the Quant fund space also continued to put pressure on our retention rate.

Approximately 50% of the cancellations in equity analytics recorded during the quarter were a function of budget constraints. 30% were due to firm or fund closures. 10% was due to price reductions. And the remaining 10% was split between clients switching from switching from Aegis to BarraOne and clients seeking other alternative product solutions either from competitors or through internal solutions. The cancellations were concentrated with about 40% coming from five clients. The run rate for Barra Aegis declined 3.3% from Q2 to $79 million, and the run rate for Equity Risk Data sold directly and through vendors declined 5.4% to $40 million from Q2.

As discussed on prior calls, our heavy investment spending in equity analytics over the last two years has started to produce what will likely continue and accelerate in terms of number of new product introductions and number of new clients. During the quarter, we released the GEM2 history data, and we released a new European equity risk model, and we expect to launch a new Canadian model in Q4. It is very likely that the number of products introduced in 2010 will accelerate both in terms of new models and in terms of enhancements to our software.

The multi-asset class analytics business run rate increased 4.1% sequentially. The run rate for BarraOne increased 8.8% sequentially to $32 million. Q3 was our third best quarter ever in terms of new sales, reflecting the loosening of client budgets and the more rapid closure of deals in the pipeline for BarraOne. We saw notable strength among asset owners in both the Americas and in Europe. Our core BarraOne risk product continues to sell well, and we saw good growth in our performance product during the quarter, and importantly, we closed our first managed services contract. Our core retention rate declined to 77.5% from 94%, reflecting primarily consolidation in various areas of our client base and hedge fund segment weakness.

Let me now turn it over to Mike for some other financial highlights.

Thank you, Henry. Rather than go through all of the quarterly numbers that are in the press release I will highlight a few areas for you. The first area is our asset-based fee business. In this business the AUM and equity ETFs linked to our indices increased $23 billion or 13% during the third quarter. We estimate that $23 billion increase in AUM is split between $20 billion from market appreciation and $3 billion for inflows. Approximately $0.6 billion of the $3 billion of inflows was into our US broad market index ETFs, and the balance was spread across a wide range of ETFs. There were outflows of $1.4 billion in both EFA and emerging markets.

Our overall market share of equity ETFs globally increased to 28% on September 1st from 27% on June 1st. Our market share of US listed ETFs remained at 32% and increased to 26% from 25% in the European listed market. We ended the quarter with 253 ETFs linked to MSCI indices, which is an increase of 31 during the quarter, with the new listings occurring in the Americas and Europe. The top 25 ETFs had a combined AUM of $148 billion and 37 ETFs had AUM balances greater than $1 billion.

Barclays accounted for approximately 60% of the AUM linked to our indices at the end August which was down from 64% in August 2008. With the increase in ETF AUM, Barclays represented 11% of revenues in Q3. The AUM balance in our ETFs, as Henry described, at the end of September, was $215 billion. At the end of Q3 our ETF run rate was $66.7 million, and our weighted average fee remained around 3.2 basis points excluding minimum fees.

The second point here I want to cover is, as we disclosed in our earnings release, we recorded a one-time revenue adjustment which reduced revenues by $3.3 million in the quarter. The revenue adjustment was made to correct how deferred revenues were being amortized in foreign entities that transact in currencies different from their functional currency. GAAP requires the exchange rate in effect at the time the contract is first recorded to be used to amortize the deferred revenue balance. We had been using the current rate to amortize deferred revenues in these entities.

The issue was most pronounced in our Barra UK and MSCI Japan entities. On the Barra UK entity, the British pound, which is the functional currency, weakened over the past several years relative to the originating currencies of euros and US dollars. As we amortized deferred revenue using the then current exchange rates for the euro and the US dollar, versus the British pound, we over amortized deferred revenues by approximately $4 million with respect to prior period. The impact can be largely seen in equity analytics and multi-asset class.

The reverse was the case in our MSCI Japanese entity where the yen, which is the functional currency, appreciated over the past several years relative to the US dollar. By using the then current exchange rates, deferred revenue was under amortized by $700,000 for prior period. The $4 million over amortization netted with the $700,000 under amortization results in a $3.3 million reduction to the current quarter's revenues. The impact of the over or under amortization of deferred revenues was not material to any prior reporting period.

Excluding the $3.3 million cumulative adjustment, revenues increased $1.8 million or 1.6% to $112.2 million compared to the same period in 2008 versus the 1.4% decline on a reported basis. The adjustment was largely recorded in equity analytics and multi-asset class. Excluding the adjustments, revenues for equity analytics were down 9.3% versus the 13.4% decline on a reported basis, and for multi-asset class were up 13.2% versus the 12.4% decline on a reported basis. Adjusted EBITDA, excluding the $3.3 million cumulative revenue adjustment, was $57.3 million.

So in summary this $3.3 million adjustment flowed through to adjusted EBITDA on a dollar for dollar basis and to net income and cash earnings on an after-tax basis.

Total operating expenses for the quarter were down 2.5% to $71 million, compared to the third quarter of 2008. If we exclude founders grant expense and depreciation and amortization expenses, our adjusted EBITDA expenses, or our cash expenses were down 7% to $55 million, comprised of a small decrease in compensation expense and a 19% decrease in non compensation expense. In managing our compensation expense, we have realized significant benefits from our emerging market center strategy. For example, despite our headcount being up 17% year-over-year, the run rate of our compensation expense at the end of the third quarter was up only 4% year-over-year because the average compensation cost per person was down 11%.

The 19% decline in non compensation expense, excluding depreciation and amortization, reflects the elimination of the Morgan Stanley allocation, as well as lower professional services fees and travel expenses, partially offset by increases in IT infrastructure costs. As noted in our second quarter call, we completed our physical separation with Morgan Stanley, and as such, we did not incur any allocation expense from Morgan Stanley in the third quarter.

Net income on a reported basis increased 10.8% to $20.9 million from the third quarter 2008. The increase in reported net income primarily reflects lower operating expenses. Our diluted EPS on a reported basis was $0.20, an increase from $0.19 in Q3 2008 reflecting higher net income. Our cash earnings for Q3 were $29 million or $0.28 on a diluted per share basis compared to $27 million or $0.26 per diluted share for Q3 2008. We calculate cash earnings of $29.2 million by taking reported net income of $20.9 million and adding back the after-tax impact of the founders grant of $4.3 million and the after tax impact of amortization of intangibles of $4 million.

And let me spend a few minutes on our balance sheet. As of August 31st, we had cash and short-term investments of approximately $362 million, which includes $293 million in US treasury securities with maturity dates within 12 months. This figure represents an increase of nearly $44 million from our balance sheet on May 31st and is after approximately $2 million of CapEx and $5.5 million in debt repayments. For all of 2009, we still expect the capital expenditures that will show on our cash flow statement for the full year to be approximately $16 million. Within the next month, we plan to settle most of our $33 million payable to Morgan Stanley.

Other cash commitments for the year include scheduled debt repayments under our credit facility which are approximately $5.5 million per quarter. Next year, our acquired principal debt repayments increase to $10.5 million per quarter. Our debt outstanding at the end of the third quarter was $385 million, and our credit ratios remain significantly better than required under our credit agreements.

For the quarter ended August 31st, 2009, our effective all-in interest rate, including the swap, was approximately 4.3%. For Q4, we expect our effective all-in interest rate to be approximately 4.2% based on LIBOR rates at the end of August. And during the third quarter 2009, our accounts receivable balance was $77 million compared to $86 million at the end of Q3 2008. The decline reflect our focus on collections.

With that we'll turn the call back over to the Operator and be happy to take your questions.

Thank you. Our question-and-answer session will be conducted electronically. (Operator Instructions) We'll take our first question from James Kissane with Bank of America Merrill Lynch.

Hi Henry and Mike. Henry, I know you said it's probably too early to say, but what's your sense in terms of the retention? Do you think you've seen the bottom, or will you see it this quarter?

Jim, it's hard to say at this point. And interestingly, even though we're partially focused on the retention rate, we are much more focused on the net effect of sales minus cancels, because, as I said, there is a little bit of a tug-of-war going on in which there are areas of our client base that are expanding, and therefore our sales are continuing to increase. And we're cautiously optimistic that that will continue, that improvement will continue. And then on the other hand, the cancels, for another part of our client base, that are putting pressure on the retention rate. So bear in mind that retention rates only take into account the cancels compared to the run rate, but don't take into account the new sales activity all in. So, therefore, I think in general what we're trying to do is basically make as much effort as possible to continue to improve the rate of new sales to mitigate the cancels, because we think that over time, as the markets improve, the high levels of cancels that we have seen will abate.

Okay, great. And related, just the outlook on pricing. It sounds like you're passing through price increases, but at the same time there might be a little more discounting going on in the renewals. So can you put the net of that all together in the outlook?

It is very different by product area. And that's very critical to follow. On equity indices, we continue to pass on moderate price increases that are going to ripple through the run rate and ripple through the financial statements as the renewal of contracts come and as those new levels of products get amortized in our financial statements. On equity portfolio analytics, over the last two quarters, we have made an effort to retain the clients, to accommodate clients that are in trouble, so that they continue to use our products as opposed to abandon them. So we have been more flexible in pricing terms to some of those clients. That is now abating. We are doing less of that, and we're hoping that will continue in the next few quarters as the crisis abates to some of those people. On equity portfolio analytics, our risk business, prices is pretty much the same. We're holding it pretty firm, but there's no price increase or price accommodation necessarily.

Perfect, thanks for the color.

We'll take our next question from David Scharf with JMP Securities.

Good morning, Henry. Couple things. One, just noting the concentration of the cancellations within portfolio analytics, I think you said 40% or 50% were just five clients. Is that typical of what you see? Is there typically that kind of concentration of the Aegis exposure?

On equity analytics, you're right, 40% of our cancels in the quarter came from five clients. A couple of those cancels were large, $1.5 million to $2 million per client. And it was not just Aegis, but also equity risk data sold directly to the client to be used with their own internal system. So I think what we're seeing is a little bit of a phenomenon that some large clients that have been under duress that have been using our products extensively have cut back, either partially or totally, in some cases broker dealers that have closed down proprietary trading desks that are basically hedge funds, hedge funds that were used in our product and services. So that's on one hand.

On the other hand, there is quite a significant number of clients that is broadbased across asset managers and hedge funds and the like, that have been reducing the level of services from us. So the story is yes, some level of client concentration, but there is also a long tail of a lot of clients that have been reducing either partially or totally across. So that's what's happening there.

But as I said in my remarks, even though the level of cancels on equity analytics have increased and reach $10-plus million out of the $17 million in total cancels, we're seeing also a strengthening of sales activity around the world, particularly around the introduction of new products, like the GEM2 product that we introduced at the end of last year, the new European risk model. And hopefully as we introduce new products in the balance of the year in 2010, the sales will continue to pick up, and at some point will cancel out the high levels of cancels.

Looking forward into Q4, if I recall from last year, during your January earnings call, there was quite a bit of focus on the fact that it was a November fiscal year, that it didn't capture the end of the calendar year, yet I think you maintained that your fiscal Q4 is still where you see the largest amount of subscriptions up for renewal. Just trying to get a sense with potentially further headcount reductions and so on and so forth how we ought to think about this quarter in terms of what percentage of portfolio analytics run rate typically comes up for renewal. Is it going to be a similar pattern as the last few years?

I think the contract renewing in December is about 18%. Equity analytics is about 24% in December. Equity indices is about 16%, on multi-asset classes about 9%. I don't have the figures in the October, November, December. That's just the month of December, as an example. I think what we have going here, our pipeline of cancels, particularly in equity analytics, in Q4, is not very large. But having said that -- and I want to emphasize this -- having said that, I think there will be a phenomenon as to describe it that those clients who were paralyzed during the crisis in the fourth quarter, fourth calendar quarter of '08, and didn't get a chance to put out cancellation notice, may end up doing so. But so far, the level of visibility for us is not very high because we try to go around and try to fill up the pipeline of cancels, and it is not very large. But again, we are forecasting ourselves that it may potentially be another quarter of high cancels, but at this point the visibility is low, and we really don't know how it's going to turn out.

Perfect, thank you very much.

We'll take our next question from John Neff with William Blair.

Thank you. Just to be clear, because there is a seasonal pattern to these retention and renewal rates, is the pressure you're anticipating in the fourth quarter due to the normal seasonal pattern of retention rates and how they're calculated, or is the pressure in addition to that pattern?

John, it is a weird period, it is a weird quarter, because typically -- the answer is probably both. There is a seasonal component, and there is a cyclical component in the current downturn. But it is a weird period, because it is conceivable that this quarter behaves differently seasonally than they have behaved in the prior years. We for sure saw that in the first, second, and third quarter. But the short answer is that yes, there will be seasonal impact because there clearly is a larger renewal rate of contracts in the fourth quarter. There will be that impact of those clients who didn't get a chance to reduce services in the fourth quarter of '08, during the beginning of the crisis. And that will be the cyclical component to that.

Okay. And then, Mike, I think you mentioned the foreign exchange adjustment hit multi-asset class and equity portfolio analytics in particular. My question is on the multi-asset class. The revenue was light, and that's a function in part of the FX impact, but the run rate there at the end of the quarter was good, and that continues to accelerate year-over-year. So just wondering if you coul comment on those dynamics. Then also I noted MAC retention was also low in 2Q '08, and, again, you mentioned that that number tends to be prone to some large cancels. So I just was wondering how you're feeling about that business in general, given the run rate.

Okay. Two parts, I'll answer the first part, then Henry maybe can expand on the run rate. But the revenue adjustment that we recorded to accounting revenues, GAAP numbers in our financial statements had no impact on run rates. We mark to market our run rates at the end of the quarter based on the relationship between the non dollar denominated run rate and the US dollar. So at the end of August, we did that, compared to where we were at the end of May. So that adjustment had no impact on the run rate. And as you correctly mentioned, the run rate in that business continues in the current market environment to grow nicely at up over 4% sequentially, but the accounting revenues were down purely as a result of that revenue adjustment we talked about.

Yes, your comment, John is correct, which is we have seen fairly good, we saw a fairly good quarter in Q3 for BarraOne. For all of multi-asset class, we continue to push those clients who are still in total risk to determine whether they want to move to BarraOne or they want to move to another system, and there was one client that migrated. So we'll continue to see cancels out of total risk, because that's what we're trying to push people to do.

The pipeline, as I said in my comments on BarraOne, remains very healthy. The risk management remains a fairly topical product for our clients worldwide. We have seen strength across the board between asset owners, asset managers, and across the regions, the US, EMEA, we've seen strength in Japan and non-Japan Asia for BarraOne to the point which we're trying to expand the number of salespeople that are calling on this product. We're trying to expand the type of services that we're providing, not only selling the product on an ASP basis, but providing managed services, and that is basically instead of clients doing the uploading and downloading of data and creating the reports and all of that, they tell us what to do and we automate it for them. Very healthy pipeline.

Let me also take an opportunity to mention that when you really analyze our business, there's no question that there are continued effects from the crisis in all product categories. But the significant relative weakness that we currently have in our overall business is equity analytics, in terms of level of sales increasing but not to the point that we're comfortable with, in terms of cancellations being pretty high. As I said $10 million out of the $17 million or so are coming from equity analytics. And, as I said, we're taking efforts to mitigate that. But when you look at the performance of equity indices and multi-asset class, and FDA and the like, in the third quarter, it was pretty robust, and it continues to look that way in this quarter.

Thank you. If you have time for one more, a quick one on the sales force would be, if you could possibly just remind us of the importance of retention rates to their incentive comp payouts. Thank you very much.

Yes, in terms of our salespeople, currently, as we have done in the past, we evaluate their performance on the basis of net new sales, and that is what we call gross sales minus the cancels in their client base. Clearly in this period of high cancels, many of them have complained to us that they have a great deal of power in figuring out how to sell to a new client, but there's almost nothing they can do to mitigate consistently the cancels. It's an internal decision by a client. But we continue to motivate them on a net new sales basis. We made a little bit of minor adjustments around that, but that's still the impact in the way we do it. And the reason is that we want them to also work pretty hard in figuring out ways how to mitigate the cancels by maybe swapping the product with another product, keeping the budget, giving them another product, providing new services, providing creative ways, in some cases when we've accommodated clients in equity analytics for price reductions to set up the contract in a way that at the renewal it comes back to the original price, et cetera, et cetera. So we're continuing to provide incentives for them to help with that.

Thanks very much.

(Operator Instructions) We'll take our next question from Andrew Fones with UBS.

Thank you. I had a question regarding the seasonality. You started mentioning Q4. You mentioned retention but could you also discuss client addition, and the seasonality you might expect, and can you base that in terms of your comment earlier about ETF fees leading the sub business.

Andrew, this is Mike Neborak, there was difficulty in hearing you. We just couldn't make out that question.

Sorry, hopefully you'll hear this more clearly. I was wondering if you could talk about the seasonality of client additions in Q4, like your comment about the ETF business leading the sub business. Thanks.

Okay. I think we understood the question, but let's take a crack at an answer, and if we don't, then we can come back to it. Let me address the ETF fees and Make can talk about the seasonality of the subscription business. As we have said in a number of calls, and in a number of client conferences and individual meetings and the like, we have two businesses. There are different ways to look at our business, but one way to look at it is that we have two business models. One is a recurring subscription business that has fixed fees attached to them, and another one, which is a recurring business but has variable fees attached to them. Clearly the latter, in the latter category, the majority of that is the (inaudible) fees, but it's not the only part. We have a number of institutional asset managers and retail asset managers that pay us on AUM as well in their mutual funds or in their institutional assets. But the majority of that category is a (inaudible) fund. Since it is all equity, and equities are a leading indicator of economic activity in the future, that business is a leading business in terms of revenues and the like.

The subscription business is a business that keeps going, even when there is a crisis and even when there is a downturn, a cyclical downturn, because there is a lag effect as to when people restructure or unsubscribe or reduce certain products and cancels and the like. As we saw, going into Q4 of last year and into Q1 of this year, that subscription business kept going and growing, even though we were in the midst of the worst financial crisis in our generation. But eventually it catches up, and that is what we call the lagging effect. And we began to see a little bit of that in Q 1, then in Q2, and now obviously the brunt of that catching up process on a net basis, gross sales minus cancel has been in Q3.

But again, it is important to differentiate the sales versus the cancels, in the overall component of the business to see what is happening, because they obey to different drivers as the cycle comes down and goes up. But in an aggregate basis, you have a lagging business that is still bottoming, which is a subscription business, and you have a leading business which is the ETF fees which is now having almost record AUM, and more than record revenues, run rate, revenues, coming our way.

And then, Andrews, in terms of seasonality, which I think is your other question, regarding our sales and new clients, historically, in terms of our addition of new clients quarter to quarter, there really is no seasonal pattern. The fourth quarter can have as many new clients as the second quarter. And this is historically going back the last three, four years. In terms of an aggregate gross level of recurring sales, the fourth quarter really, if you go back historically, doesn't show any seasonal pattern as being the worst. Typically in the past, our first quarter had always been our best quarter in terms of recurring sales. That was not the case this year. As Henry mentioned in the first quarter of 2009, we had about $10 million of recurring sales and in the second quarter it was $14 million, and then most recently here in the third quarter was $16 million. So some of those seasonal patterns might not hold. But if you go back historically, there is really not any identifiable seasonality related to fourth quarter new sales generation or new client generation.

Okay, thank you.

(Operator Instructions) We'll take a follow-up from John Neff with William Blair.

Thank you, I'll try to be quick. Could you possibly provide, if you haven't, gross client adds? I think it was up 37 sequentially in the quarter, but curious what the gross number was. Another quick question would be, I think, Mike, you mentioned ETF flows or ETF asset levels at the end of September $215 billion. How much of that was flows? Then a third question, quick one here, but it's interesting to note how the US broad market ETF, that that is now a top three ETF in terms of assets. Is there any reflection there on inroads you might be making in terms of US market benchmarking or mandates? Thanks very much.

Okay, in terms of the $215 billion dollar balance at the end of September, John, I do not really have the breakout in terms of the increase from where we were at the end of August between flows and market appreciation. And then in terms of your question, I think what you were referring to is we had a net client change in the third quarter of 17, and the components were 86 new clients and 69 lost clients, which nets to the 17 addition.

Now, in terms of the last question, John, in the ETF, but also in terms of benchmarks, yes, we are pretty focused on a strategy to penetrate the business of US equity index benchmarks for US investors. Bear in mind that the MSCI USA indices are the benchmark for the non US investors. So our Japanese investor globally, MCSI USA is the benchmark, similarly to Australia or Chinese or European, et cetera.

But in terms of US investors, being benchmarked to US assets, we traditionally haven't been a large player there, and we have a concerted strategy to develop that market in two categories. One in the change credit fund, and the Vanguard products in ETF are making inroads into that, and that broad market index is a reflection of that. Secondly, and very importantly, is that in the last few quarters, we also have made inroads into convincing pension funds in the US to be benchmarked to either the old country world index or the old country world index investable, which includes small cap, and that would, therefore, include the US portion of the portfolio. So a handful of pension funds have either started that process or have made a decision to have as an overall policy benchmark for the pension fund the MSCI after we benchmark, not excluding the US.

So as we continue with that, the net effect of that will be that both our developed market and our emerging market business will continue to grow. Hopefully for those who pick the broadest benchmark, our global and small cap business will continue to grow. But very importantly, we'll end up picking up a significant amount of assets on the benchmark in the US equity portion of the entire portfolio. So more to come on that.

Thanks very much.

This will conclude today's question and answer session. At this time I would like to turn the call back over to Mr. Fernandez for any additional or closing remarks.

Thank you, Operator. So let me just summarize again where we are in the third quarter. We had fairly excellent results in our changed credit fund business. We had fairly significant improvement both in sales and in some cases cancels for three of our product categories -- equity indices, multi-asset class, in some cases, and our energy and commodity analytics.

Clearly an area of relative weakness for us is the equity portfolio analytics business in which sales have recovered quarter over quarter, sequentially, but the level of cancels have more than eaten into the new sales. We're taking steps to do two things there -- to continue to sell aggressively and look for new product introductions in the balance of the year in 2010 to continue to sell, and work with our clients in ways that we can mitigate those cancels. We are cautiously optimistic that over time, as the operating environment improves altogether, that many of those cancels in equity analytics will start to abate, but we really don't know when all of that is going to happen yet. And as we have said in the past, this is a business of very strong levels of incremental profitability. We're managing our expense base extremely tightly. We are making huge investments and in the way of headcount by adding people around the world in our emerging market centers, and over time that will translate into acceleration of revenues and continued improvement in the profitability of the business. Thank you, and thank you for joining us today.

This will conclude today's conference. Thank you for your participation.