Welcome to the Marsh & McLennan Companies conference call. Today's call is being recorded. Third quarter 2019 financial results and supplemental information were issued earlier this morning. They are available on the company's website at www.mmc.com. Please note that remarks made today may include forward-looking statements. Forward-looking statements are subject to risk and uncertainties, and a variety of factors may cause actual results to differ materially from those contemplated by such statements. For a more detailed discussion of those factors, please refer to our earnings release for this quarter and to our most recent SEC filings, including our most recent Form 10-K, all of which are available on the MMC website. During the call today, we may also discuss certain non-GAAP financial measures. For a reconciliation of these measures to the most closely comparable GAAP measures, please refer to the schedule in today's earnings release.
I'll now turn this over to Dan Glaser, President and CEO of Marsh & McLennan Companies. Please go ahead.
Thank you. Good morning. Thank you for joining us to discuss our third quarter results reported earlier today. I'm Dan Glaser, President and CEO of Marsh & McLennan. Joining me on the call today is Mark McGivney, our CFO, and the CEOs of our businesses, John Doyle of Marsh, Martine Ferland of Mercer, and Scott McDonald of Oliver Wyman. Unfortunately, Peter Hearn of Guy Carpenter is sick with the flu and cannot join us today. Also with us this morning is Sarah DeWitt, Head of Investor Relations. We are pleased with our third quarter and year-to-date results. We produced excellent top-line growth in the quarter with underlying revenue growth across both segments, our adjusted EPS growth is tracking well with our expectations. We are generating strong results while continuing to make progress on the JLT integration, overall, the combination is progressing nicely.
Our clients are beginning to see the power of our combined firm, we are demonstrating that we are better positioned to help them with their greatest challenges through our enhanced talent, capabilities, and geographic footprint. We are meeting our expectations in terms of our financial targets, revenue growth, and colleague retention, our businesses are coming together. Co-locating our teams is a critical part of our integration because it enables the cultures to coalesce as well as results in increased collaboration. To date, we moved over 8,000 colleagues, by the end of 2019, nearly all colleagues will be sitting side by side in their respective areas of operation globally. We have also undertaken significant work to integrate our platforms. We now have nearly all of the JLT colleagues on our HR system and have integrated financial reporting as well as our email environments.
This was possible because of investments we have made over the years to harmonize our systems and strengthen our infrastructure. Our progress to date makes me confident we will comfortably exceed our $250 million savings target over three years. As we have said from the beginning, our combination with JLT is about growth and the benefits this acquisition brings to clients, colleagues, and shareholders. More and more, we are working as one team for our clients. For example, colleagues from several Marsh-JLT Specialty practices across multiple countries came together to provide a customized solution for a company managing a complex hydroelectric project. Through global collaboration and our augmented capabilities, we handled all of the project's insurance requirements. In Guy Carpenter, our cyber insurance team won an RFP for a large U.S. regional carrier by demonstrating the strength of the combined organization.
In Indonesia, a large manufacturer had been engaging carriers directly for their EH&B services, resulting in a sub-optimized outcome for the client. Mercer Marsh Benefits, which includes former JLT colleagues, reviewed the existing benefits and recommended critical changes to drive a better outcome. This resulted in new business for MMB, lower premium costs for the client, and a more optimized solution for our client's employees. We also continue to capitalize on opportunities to serve clients across our businesses. For example, we recently leveraged our collective strengths to create a compelling global value proposition for an Asia Pacific transportation company who is upgrading their infrastructure to a sustainable modern transportation network. We won the mandate by leveraging Marsh-JLT Specialty's construction and marine expertise, Oliver Wyman's transportation and public sector expertise, and Guy Carpenter's risk transfer and alternative capital capabilities.
During the process, we also introduced MMB, who was appointed to manage the employee benefit portfolio. Overall, I am proud of our company's progress, impressed by the hard work of our colleagues, and grateful for the ongoing support of our clients. Let me spend a moment on current P&C insurance pricing trends. Pricing is firming across a wide range of geographies and lines. The Marsh Global Insurance Market Index saw an increase of nearly 8% in the third quarter, compared with 6% in the second quarter and 3% in the first. Global property insurance and financial and professional lines saw the highest average renewal rate increases at 10% and 14%, respectively. Casualty rates are up 1% on average, but mixed by line, with excess casualty rates rising and workers' comp down.
Note that the Marsh index skews to larger risks, which are seeing higher increases, although middle market and small commercial insurance rates are up in certain geographies. Turning to reinsurance. In the property catastrophe market, we have seen reinsurance rates increase throughout 2019 as a function of loss activity, geography, and exposure increases. Overall, we are focused on driving the best outcome for our clients. In markets like these, our capabilities and expertise shine through and become even more critical. Now let me turn to our third quarter financial performance. We delivered excellent results in the quarter, with underlying revenue growth across both Risk and Insurance Services and Consulting. We are executing well, both delivering for clients while working through the integration of JLT. Total revenue was $4 billion, up 13%, or 5% on an underlying basis.
Adjusted operating income increased 10% versus a year ago to $585 million, and the adjusted operating margin increased 10 basis points to 16.9%. Adjusted earnings per share fell 1% versus a year ago to $0.77, reflecting seasonality at JLT. Year to date, adjusted EPS increased 6% to $3.47. In Risk and Insurance Services, third quarter revenue was $2.2 billion, an increase of 18%. Underlying revenue growth was strong at 6% in the quarter. This reflects 5% growth in Marsh and 11% at Guy Carpenter, which was helped by a true-up of a multiyear contract, which Mark will discuss in more detail. RIS adjusted operating income increased 11% to $313 million. The adjusted operating margin was 17.4%, reflecting JLT seasonality. In Consulting, third quarter revenue was $1.8 billion, up 8% compared with a year ago.
Underlying revenue was up 4% for the quarter, driven by underlying growth of 7% in Oliver Wyman and 3% in Mercer. Consulting adjusted operating income grew 9%, and the adjusted margin expanded 50 basis points versus a year ago. In summary, we are pleased with our third quarter results and our progress integrating JLT. For 2019, we expect a solid first year as a combined company, underlying revenue growth in the 3%-5% range, margin expansion, and solid growth in adjusted EPS. With that, let me turn it over to Mark for a more detailed review of our results.
Thank you, Danny. Good morning. We had an excellent quarter. Overall revenue grew 13% to $4 billion, reflecting the meaningful step forward JLT represents for our firm. We generated strong underlying revenue growth of 5%, with 6% in RIS and 4% in Consulting. Remember that underlying revenue growth in our communications and disclosures includes JLT. Operating income in the quarter was $467 million, while adjusted operating income increased 10% to $585 million. Overall, our adjusted operating margin increased 10 basis points in the quarter to 16.9%, a good result given the seasonality at JLT. Adjusted EPS increased to $0.59 in the quarter, and adjusted EPS was $0.77. For the first nine months of 2019, total revenue growth was 10%, with underlying growth of 4%. Our adjusted operating income grew 13%, our adjusted operating margin increased 110 basis points to 22%, and our adjusted EPS increased 6% to $3.47.
Overall, our year-to-date performance leaves us well-positioned for a solid first year with JLT. In Risk and Insurance Services, third quarter revenue grew 18% to $2.2 billion, with underlying growth of 6%. RIS has now produced 5% or higher underlying growth in five of the last six quarters. Adjusted operating income increased 11% to $313 million, and the adjusted margin decreased 30 basis points to 17.4%. For the first nine months of the year, revenue was $7.2 billion, with total revenue growth of 14% and underlying growth of 4%. Adjusted operating income for the first nine months of the year was up 12% to $1.7 billion. At Marsh, revenue in the quarter was $1.9 billion, with underlying growth of 5%, representing another strong quarter of growth. U.S. and Canada grew 6% on an underlying basis in the quarter.
This marks the sixth consecutive quarter that U.S. and Canada delivered 5% or higher underlying growth. International underlying growth was 3%, with Asia-Pacific up 7%, EMEA up 2%, and Latin America down 1%. For the first nine months, revenue at Marsh was $5.8 billion, with underlying growth of 4%. U.S. and Canada was up 5%, while international was up 3%. Guy Carpenter's revenue was $273 million in the quarter, with underlying growth of 11%. Strength in the quarter was due to solid operating performance and a true-up of a multi-year contract. Excluding the multi-year contract adjustment, underlying growth was mid-single digits in the quarter. The first nine months of the year, Guy Carpenter's revenue was $1.3 billion with 4% underlying growth. In the consulting segment, revenue in the quarter was up 8% to $1.8 billion, with underlying growth of 4%.
Adjusted operating income increased 9% to $320 million, and the adjusted margin increased 50 basis points to 18.7%. Consulting's underlying revenue growth for the first nine months of 2019 was 4%, with consolidated revenue of $5.3 billion. Adjusted operating income for the first nine months of the year was up 13% to $916 million. Mercer's revenue was $1.3 billion in the quarter, with underlying growth of 3%. Health grew 7% in the quarter, the best underlying growth since the first quarter of 2018. Career underlying growth was solid at 5%, and wealth underlying growth was flat, with mid-single-digit growth in investment management offset by a low single-digit decline in our defined benefit business. Our delegated asset management business continues to show strong growth, with assets under management increasing to $290 billion. For the first nine months of the year, revenue at Mercer was $3.7 billion, with 2% underlying growth.
Oliver Wyman's revenue was $505 million in the quarter, with strong underlying growth of 7%. This was the fifth straight quarter of 7% or higher underlying growth at Oliver Wyman. Growth was solid across most regions, with our financial services and health and life sciences practices showing particular strength. First nine months of the year, revenue was $1.6 billion, with 9% underlying growth. Turning to corporate, adjusted corporate expense was $48 million in the quarter. Based on our current outlook, we expect approximately $50 million in the fourth quarter. Turning to the JLT integration. This is our second full quarter post-closing, and I couldn't be happier with the progress we've made to date. Dan said the integration is going well from an operational perspective, and we are on track with our financial targets.
We continue to expect the transaction will be modestly dilutive to each adjusted EPS in the first year, break even in year two, and be accretive in year three. We are ahead of schedule on cost savings and associated restructuring charges. We expect to exceed the $250 million of run rate savings and the $375 million of cost to achieve those savings. We currently plan to provide an update on our outlook for cost savings on our fourth quarter earnings call. During the quarter, we incurred $133 million of interest expense. We expect around $130 million of interest expense in the fourth quarter. In the third quarter, we recorded $84 million of amortization, which includes the JLT-related amount, as well as amortization from other transactions. Our overall view of amortization related to the JLT transaction is now $163 million annually.
As a result, we expect total deal-related amortization in the fourth quarter will be about $93 million. In aggregate, the financial impact of the JLT transaction is tracking well with our initial expectations. As we look to the fourth quarter, there are a few things to keep in mind. Marsh faces a tough underlying revenue comparison to a year ago and a continued impact on new business in JLT's seasonally strongest quarter. As a result, we expect Marsh's underlying growth to moderate in the fourth quarter, but still be solid for the year. In addition, Oliver Wyman tends to be our most volatile business quarter-to-quarter, as we've discussed in the past, and our current outlook calls for a pullback in the fourth quarter. Overall, however, Oliver Wyman is on track for a strong year.
In terms of the fourth quarter adjusted operating margin, we expect strong margin expansion that is moderately higher than the pace year-to-date, which reflects expense savings and a seasonally strong quarter at JLT, partially offset by the sale of JLT's aerospace business, which generated nearly all of its earnings in the fourth quarter. Turning back to the third quarter, we reported $118 million of noteworthy items, mostly related to the JLT acquisition. Included in this total are $77 million of JLT integration costs, the largest category of which is severance, $21 million of JLT acquisition-related costs, and $12 million of other restructuring costs, mainly related to Mercer's program. Turning to investment income. On an adjusted basis, we had $3 million of investment income in the quarter, and we continue to expect the contribution from investment income for the balance of 2019 will be immaterial.
On a GAAP basis, investment income was $7 million in the quarter. Foreign exchange was a slight benefit to adjusted EPS in the quarter. Assuming exchange rates remain at current levels, we expect FX to be a $0.02 per share headwind in the fourth quarter. Our adjusted effective tax rate in the third quarter was 25%, compared with 25.3% in the third quarter last year. Through the first nine months of the year, our adjusted effective tax rate was 24.3%, compared with 24.5% last year.
Based on the current environment, we continue to expect a tax rate between 25% and 26% for 2019, excluding discrete items. Total debt at the end of the third quarter was $12.6 billion, or $12.2 billion excluding commercial paper. In September, we repaid $300 million of senior notes that matured, consistent with our de-leveraging plan. Our next scheduled debt maturity is in March 2020, when $500 million of senior notes will mature. In the third quarter, we repurchased 2.1 million shares of our stock for $200 million. For nine months, we have repurchased 3.1 million shares for $300 million. We continue to expect to repurchase enough shares in 2019 to reduce our share count. Our cash position at the end of the third quarter was $1.2 billion.
Uses of cash in the third quarter totaled $486 million, including $53 million for acquisitions, $233 million for dividends, and $200 million for share repurchases. For the first nine months, uses of cash totaled $7.1 billion and included $6.1 billion for acquisitions, $655 million for dividends, and $300 million for share repurchases. Overall, we're on track to deliver a solid year. With that, I'm happy to turn it back to Dan.
Thank you, Mark. [ McLennan] . We're ready to start the Q&A.
Thank you. In the interest of addressing questions from as many participants as possible, we would ask the participants limit themselves to one question and one follow-up question. If you'd like to ask a question, please signal by pressing star one on your telephone keypad. If you're using a speakerphone, please make sure your mute function is switched off to allow your signal to reach our equipment. We will now take our first question from Mike Zaremski from Credit Suisse. Please go ahead.
Hey, good morning. First question on the insurance brokerage side of the business. We're increasingly hearing from some of the larger corporations and insurers that deal with, let's say, Fortune 500 space, that pricing is much higher there than, let's call it, the SME, small to mid-size space. I'm aware that Marsh's revenue contracts in that space are more fee-based. I get asked a lot from clients, how do we think about how the impact on to the broker when pricing is very high in the large account space? Maybe you can give us a sense of are you seeing that as well, that dynamic? Historically, have your fee rates increased at a similar rate to maybe the overall brokerage's organic rate of growth? Anything there would help.
Sure. I'll start, and then I'll hand over to John to give you some more information. I think the first thing to know is that we've built the business to operate across cycles. Softening cycles usually last for a long time, and there's periods of tightening, which generally are much shorter in duration. Every market is different, but we're not built or geared based upon the cycle. You're right in saying that in the large account space, it skews a little bit or a lot more to fee than to commission. You have to bear in mind between fees, we have discussions with our clients all the time with regard to the value that we're creating. Generally, in the large account space, there's always a year-over-year type of discussion. It's hard to get fees to increase.
As you can imagine in large accounts across the world, you have procurement departments and risk management experts who negotiate on their behalf. When John and I look at this overall, we were facing a mild headwind for a number of years, going back a few years ago, and maybe we have a mild tailwind today, but it's not material to the overall performance of the company. John, you want to tackle that?
Sure, Dan. As you said, I think it's a modest tailwind at the moment. I think it's really important to keep in mind, of course, that our role is to get the best outcome for the client, regardless of where we are in a pricing cycle. A good chunk of our portion of our large account business is on a fee, for sure. We also have Marsh Risk Consulting, which is a fee-based business, and our benefits business, which obviously is not subject to the P&C pricing cycle. At the moment, middle market pricing is not moving up as much as large account pricing, which we have more exposure to from a commission point of view. We're certainly working with our clients to mitigate the impact of price increases.
In a lot of cases, whether it's in large accounts or in the middle market, that may mean our clients retaining more risk. It's a challenging market in some segments. I don't consider it actually one market. It's a collection of markets. It's a great time for us with our tremendous talent. We have great talented colleagues and capabilities to create value for our clients.
Can you elaborate where it's. I guess I'll move to one more on defined benefits. You mentioned a low single-digit decline there. I believe you're in the midst of taking actions to try to improve that level going into 2020. I guess related to, given the big dip in interest rates year-to-date versus last year, does that business get maybe a pickup later in the year due to client activity kind of looking into their impacts to the pension funds from low rates?
Mike, I'll start with that and then hand over to Martine to give you more detail. Mercer has done a really nice job over several years, essentially preparing themselves to be able to deliver high-quality services to clients while we're earning a good margin on the business in a secular decline of defined benefit pension plans. That's playing itself out. We're not surprised by the low single-digit decline, those declines we saw last year as well. Martine, do you have any further to add about the impact of interest rates and that sort of thing?
Yes, of course. As Dan just said, our wealth business is now composed of investments as much as the DB core benefit consulting. Indeed, the economic conditions, and the market conditions can influence the rhythm at which the employers will decide to de-risk their DB plans. There is a little impact of that. It is a temporary measure because over time, these plans are getting closed by our clients, and we're prepared for that. Of course, also the market conditions do impact our large investments, consulting, and in particular, our OCIO business. There again, our clients maintain a very balanced portfolio and therefore the impact is mitigated because we have asset classes across equities, fixed income, and alternatives.
Thanks. Next question, please.
Next question comes from Elyse Greenspan from Wells Fargo. Please go ahead.
Thanks. Good morning. My first question is just on Guy Carpenter. I know you guys said that there was a multi-year contract that did help the organic. Ex that it would have been in the mid-single digits this quarter. That's still a nice pickup from where you guys were in the second quarter. I was hoping to get a little bit more color. Was that some new business wins? I know, Dan, you mentioned cyber retention. I think, last quarter you had mentioned that there was an impact on new business from JLT. Just trying to get a sense of if things turned around a little bit in the third quarter.
Sure. Just to be clear, I'm happy to fill in with Peter anytime he has a double-digit quarter. Yeah, you're right, it was a bit of a positive surprise to us in the third quarter. Bear in mind, it's a small revenue quarter for Guy Carpenter, so it doesn't take much for an outsized quarter in either direction. JLT is going to continue to be a drag to underlying revenue growth as Carpenter rebuilds the new business pipeline. That's something that takes multiple quarters to do. That effort is already underway. It's not going to last for years. It's temporary. In the meantime, Carpenter's doing a great job managing expenses to deliver earnings growth. We mentioned the multi-year contract true-up. We called it out not because it's unusual, because it was large in a small quarter.
That's why we raised it. It's a normal course of business type of situation.
Okay, thanks. In terms of the margins within RIS, I know JLT was a drag this quarter, but the margins still seem to be trending better than I would have expected. If you could give a little bit of color. I know you guys said savings running better than planned. Were the savings directionally picked up Q3 to Q2, maybe it's some level of improved organic? I'm just trying to get a sense of what's driving the still pretty strong margins, especially in a quarter where JLT loses money.
Yeah. Overall, you know that Marsh & McLennan had a pretty materially higher margin than JLT. Both very high-quality organizations. Marsh & McLennan benefited, I think, more from some of the scale, and broader heft that we had as an organization. Every intention was to convert JLT within Marsh & McLennan to Marsh & McLennan style margins and look to grow from there. We're pleased with the margin expansion that we're having this year. We expected it, as we talked going into the year, we said that we expected to have margin expansion. It really is about we're bringing in more revenue, and we're doing it in a more efficient way.
The notion that we feel comfortable at this stage, which is still really early stages of coming together as a single organization, that we will comfortably exceed the $250 million of expense savings that we cited last September when the deal was announced. As we've mentioned before, we believe the majority of those savings are going to drop to the bottom line, we're seeing that right now. As you can expect, even though JLT, the acquisition itself was not an expense synergy play, it's all about growth. We feel very happy with the level of talent and capabilities, the opportunities for colleagues that are developing. We expect revenues and earnings over time to be better as a result of this combination.
Having said all that, we're a well-run organization, when we see areas for capturing efficiencies, areas like real estate, technology, functions like finance, legal, HR, risk, compliance, all give us areas to where we believe we can be more efficient together. What we'll do, as Mark McGivney was mentioning in his script, we'll give a more fulsome update next quarter about where we see the three-year expenses and cost of achieving those expense saves. We'll do a review internally, and we'll let you know next quarter. Next question, please.
We will now take the next question from Paul Newsome from Sandler O'Neill + Partners. Please go ahead.
Good morning. Couple of follow-ups. The issues with the comparison for organic growth in Marsh in the fourth quarter, would that also extend into the first quarter next year, given how strong the organic growth was in the first quarter, or is that not the right comparison? Obviously, the JLT deal makes that a little bit more complicated.
Yeah. Overall, we're pleased with the year-to-date performance of not only Marsh but RIS and the overall firm, the consulting segment as well. We believe the way to evaluate the firm, both on the top line and on the bottom line, is over multiple quarters and not looking at any one quarter. Although, having said that, we'll bask in the glory of a terrific third quarter for a little bit of time. Ultimately, 2019 is shaping up to be a solid year, especially in view of the effort of so many people within the organization who are working on integration efforts. Those are people who are legacy Marsh & McLennan and legacy JLT. We're all working together increasingly as one team to form a new organization that's quite formidable into the future.
We wanted to call out the fourth quarter just because we expect it to be more challenging. We mentioned the importance of new business and new business pipeline. Well, the fourth quarter is JLT's highest new business quarter. If you play out that logic, well, then there'll be more strain in terms of achieving what they used to achieve because of pipeline issues that developed soon after the announcement was made. Those things are all solvable. Now, whether we extend our view into next year or not, John, do you have any early read?
I think you said it well, Dan, that the fourth quarter was such a seasonally strong new business quarter for JLT, we'll see a little bit more headwind there in the fourth quarter. We're well positioned for growth in 2020. Again, if you take a longer view, we're confident about where we're positioned in the market.
Thanks. Anything else, Paul?
Just, again, really more of a modeling question. There's a fair amount of disposals happening. Obviously, you had the aerospace business. It looks like the international business was where the disposals were. How should we think about that little piece going forward when we're trying to get to our overall revenue growth estimates? Does that change with JLT?
Mark, you want to hit- Paul. The third quarter was noisy in that underlying growth schedule, there were a couple of things. There's the sale of aerospace that you mentioned. There was also the sale of a data business in Marsh last year where a gain was recorded in revenue. There's a little bit of a heavier impact. If you looked at the year-to-date schedule, you see a plus one for acquired revenue, that's probably not a bad estimate for the full year. There's actually a lot of the transactions we did this year, Lovitt & Touche and Bouchard and MMA came relatively early in the year. There's less rollover revenue just given the lower activity this year than we would typically have.
Maybe a little bit of rollover into the early part of next year, and then next year's acquired growth will really depend on the level of deal activity. Next question, please.
The next question comes from Yaron Kinar from Goldman Sachs. Please go ahead.
Hi. Good morning. First question is really, I guess, a follow-up on Elyse's question. Just trying to get a sense of what the cost save impact was in the third quarter, and maybe even year-to-date.
We're not going to go into the details of how the cost saves are breaking out per quarter. As I mentioned before, this is still really early stages. We're in the second quarter of being a combined company. We believe we have a multi-year approach to capturing the expense savings that we're identifying. We're executing on a lot of different issues right now. Clearly, with our margins being up as a company through nine months, you pretty much can say that we're achieving a lot of savings, right? It's not impacting growth because we're growing well as a firm. I don't really want to get into a quarter-by-quarter description of all the activity we're doing on the expense side.
Okay. Can you at least comment on whether if the cost saves should be expected to be more back-weighted than front-weighted?
No. I think partly is what you're seeing in margin right now. We're jumping on efficiencies and making savings as we go, and so there's not a sense of all of this is going to happen In the third year. There's many things that we can operate very quickly on. There's other things that take a little bit more time. You take something like, as an example, the consolidation of vendors or real estate or reducing costs within JLT that were geared to being a public company. Those sort of things we can move pretty quickly on. There's other areas, let's say some areas of application technology, which take a little bit more time as they go.
We have a three-year plan, and we're working through it, this is our 12th year of margin expansion, so I'm not going too far out on a limb when I say, well, I think in 2020 we're going to expand our margin, just like we did in 2019 and just like we've done for the last 12 years. So that's an indication of some of the benefits dropping to the bottom line.
Okay. Understood. My second question goes to the casualty business in Marsh. I guess one thing I struggle with is, and this isn't a Marsh issue, it's a broader industry question. I guess we're seeing rate firming excluding workers' comp, even as loss picks, excluding commercial auto, remain pretty low by historical standards. I'm assuming that's something that you as a broker see as well. Just curious as to how much pushback there is in the conversations with insurers on kind of the absence of more material loss trends. Are you actually seeing real-time loss trends picking up in claims emergence and maybe not pushing back as much?
John, you want to handle that one?
Sure. The underwriting community is clearly concerned about rising loss cost trends. Dan and I, and some of the rest of our leaders have been at Peter Hearn as well, have been at a number of different conferences where we've spent a lot of time with the markets of late. There are clearly some very large casualty losses in the markets. There's a lot of discussion, of course, about cat property, but there's some casualty cat events that are in the market that are a concern. Then there's some big headline verdicts and settlements that have underwriters concerned as well. I think we've seen loss cost trends begin to emerge more quickly in commercial auto. Pricing's been up there for a number of different quarters. Overall, casualty pricing at the moment is mixed. Work comp is down 5 GL.
Primary GL pricing is still down modestly, including in the U.S. Excess liability is up 6% versus 3% in the second quarter. That market's seeing a bit more stress at the moment. We, of course, again, are trying to get the best outcome for our clients and we'll find solutions no matter where they might be in the world. I do expect upward pricing pressure to continue throughout next year. However, I think we'll see shorter and shallower cycles than we have on average in the past. Really kind of micro cycles. Underwriters are moving quickly. They have better data, better management information. They're moving more quickly to deal with things. Capital moves very quickly too, where profit pools emerge as well. Again, I expect more shallow and shorter pricing cycles.
I think John's point is really important to emphasize here. Better management teams at underwriting firms, better data and analytics creates more of a microcycle environment where it's not necessarily broad across all sectors and all segments. That's really occurring on the reinsurance side as well. We're not seeing broad market-wide impact, even as a result of the loss creep that we've seen over the last six months coming in from prior years and some big losses this year. There is potential significant changes on individual programs based upon the individual program's characteristics like geography, performance in the past, and that sort of thing. I think the better data and analytics are making it more of a targeted approach as opposed to broad brush. Next question, please.
The next question comes from Ryan Tunis from Autonomous Research. Please go ahead.
Hi, this is Crystal Lu in for Ryan Tunis. Our first question is about the EMEA organic growth. It seems to have been improving recently. Can you just give some comments on what drove organic growth improvement there?
Go ahead, John.
Sure. I would say, first of all, overall, I was pleased with the growth in the quarter. It's 17% GAAP growth, 5% underlying growth. Mark talked about the strong, consistent growth in the United States. Internationally, we're up 3% on an underlying basis. In EMEA, we had very good results in Continental Europe. We had excellent growth in the Middle East and in Africa as well. In the U.K., our results are showing some positive momentum. As I've talked about in the past, we've made some leadership changes there. It's a challenging environment. Obviously, the U.K. economy creating a bit of a drag force at the moment. Our leadership team is really on it. I see some improving trends there for us as we look into next year.
Great. Then on the retention compensation that is coming through, can you give an idea of what kind of quarterly level we should expect on retention compensation going forward? And when we can expect these payments to be concluded? Thanks.
Yeah. In the context of acquisitions that we have done in the past, it is quite common for us to either have earn-out arrangements or to have some sort of retention payments tied to a multi-year view of how a person performs within the business. So we view that as normal. It has all been modeled within our original deal consideration. You would expect that to really be more in the first year or 18 months, and then reduce from there, because it is geared toward assisting in the transition of the organization to a new organization. So I would not expect any significant levels to persist over a many-year period. Next question, please.
The next question comes from David Styblo from Jefferies. Please go ahead.
Hi. Good morning. Thanks for the questions. Just hoping we could peel back the onion a little bit more on the cost phase, and if you can provide additional color about the increased level that you are expecting to achieve in the out-year there. Is that more of just finding more opportunities within the same subset, or are there just some new things that are emerging? To the extent that tying that to the accretion guidance, is it a situation where maybe you just do not want to comment yet on perhaps doing better than break even next year? Or perhaps should we expect that there might be an update where there could be some upside to the break-even target for next year?
Yeah. In any large organization and in any large combination, there's a series of puts and takes. I think that this is how you need to view this, in that we did not know on the day of acquisition that we would have to dispose of the aerospace business, as an example. On the other hand, we had an idea that $250 million or so of expense synergy should be achievable over a three-year period. It is the fact that we're able to achieve more than that, but at the same time, we've had some disposals that we did not anticipate. I can tell you there's probably a dozen of these puts and takes that you look forward. We end up about where we were. We're exceeding our expectations mildly on virtually every measure.
We end up with the view that it's going to be, on an adjusted EPS basis, slightly dilutive to what we otherwise would have done this year, break even next year, and accretive in year three. That's kind of where we are. The organization, for a long period of time, has become more and more efficient. We continue to see ways of becoming more efficient. We run ourselves very much to where we've got four very strong brands, but we're not a holding company just operating a business that is not cohesive. Where we are close to the client, we're commercially agile, nimble, highly segmented, highly specialized. The further you are away from the client, we're more horizontal, and we capture synergies around finance, legal, HR, and other functional costs.
We're smart about how we approach the use of service operations in places like India and Poland and Kuala Lumpur, we will continue to do that. Personally, I think we're at the early stage of being able to look at AI, machine learning, robotics as a way to increase the efficiency of the business. We can look out for several years and believe we can continue to expand margins in the business.
Great. That's helpful to frame it. Thanks for that, Dan. Maybe a follow-up for Mark real quick. Do you have maybe some initial thoughts about how we should think about capital deployment in 2020? Would it be maybe perhaps the same approach for 2019, where you'd buy enough stock to help shares come down a little bit while paying down debt and opportunistically looking for M&A, or would there be any noticeable changes from how you approached this year?
Yeah, David, going back to, as with a lot of things that we said initially, that things are playing out very close to where we initially thought they would, and that applies to capital management as well. This year has played out almost exactly as we thought. If you remember back to some of the things we said, we said the focus for capital management for the first couple of years would be de-leveraging. That in those plans, that we provided enough flexibility for return of capital as well as M&A, with the focus for M&A being on Marsh McLennan Agency. As we look out to 2020, that really still is the focus.
We're focused on de-leveraging, and that we see we've provided for some flexibility so that we'll be able to have some return of capital as well as continue to focus on M&A. The balance between those two is just going to depend on the strength of the M&A pipeline.
I think it's important to remember the way we prioritize our capital deployment is we put organic investment ahead of everything else. We put our dividend, which we believe is sacrosanct, and we've made a commitment to increase our dividend double digit each year, and we intend to continue to do that. We have said in the past that we favor acquisitions over share repurchase, and we favor share repurchase over growing cash on our balance sheet. When you look at it in that context, you have to recognize that if push came to shove, we would favor acquisitions over share repurchase. The way we're looking at our pipeline, it's very strong. We will continue to take steps to make us a stronger company. If that means that puts a little pressure sometimes on share repurchase, then so be it. Next question, please.
The next question comes from Meyer Shields from KBW. Please go ahead.
Great. Thanks. Good morning. Two very quick questions. First, can you give us a sense as to what clients are assuming for exposure unit growth as they do their 2020 planning?
That's a good question. A tough one. John?
I don't think clients are assuming unit exposure. They have their own units of exposure growth. What I would say is that in some of the more mature economies that we operate in, where we have big revenue streams, there continues to be, particularly there in the United States, modest exposure growth. Employment continues to grow, sales are growing. The number of vehicles on the street, more flat, for example, most areas of exposure management in mature markets are showing some modest growth.
Okay. Perfect.
Go ahead. Next question, Meyer.
Yeah. I was just going to ask, with regard to the Guy Carpenter true-up, does all of that adjustment fall to the bottom line, or are there offsetting expenses?
That's a great one. Mark, do you want to handle that? There's always the bonus pool, right?
Yes.
Nothing fully drops to the bottom line.
Right. Our bonus pools tend to be geared to earnings, but other than that, it really does fall into the quarter.
Okay. Thanks so much.
Sure. Next question, please.
The next question comes from Brian Meredith from UBS. Please go ahead.
Thanks. First one, just curious, LatAm, negative organic. I think it's the first time I've ever seen that out of the LatAm business. Is that related to JLT or something else happened there?
You're right in citing that. This is the first quarter of negative growth in LatAm since we started reporting LatAm in the first quarter of 2008. It's a great region for us, and it's usually neck and neck over long stretches of time with Asia as to which grows better. There's some particular issues that we're facing in LatAm in the short term. John, you want to give more color?
Sure. It is legacy JLT related. It's both new business that we've talked about kind of more broadly and some larger non-recurring items in the region. What I would say, though, is that the region absolutely remains a growth market for us. I'm really encouraged by how the teams are coming together at both Marsh and JLT. While it'll be a choppy second half for Latin America, I do expect good growth from the region in 2020.
Thanks. Great. Mark, I'm just curious, a pop in CapEx in the quarter, looked like $123 million. Was there something kind of unusual going in that? Also on that free cash flow actually was pretty strong. Was there anything unusual maybe in operating cash flow?
Mark?
Yeah, sure. Let me tackle CapEx first. It did increase. It was mainly related to timing of big real estate projects, and we can see that volatility from year to year. Actually, it will not remain at that level into the fourth quarter. Fourth quarter, we're probably looking more in line with what we saw in the fourth quarters of the last couple of years. It's just a couple of big real estate projects. Actually, the level in Q2 was a little bit lower for the same reason. In terms of cash flow growth, it was actually a really good story in the quarter. Our operating cash flow, notwithstanding all of the integration-related charges, was only down 2%. I thought that was a really good result in the context of what we're going through. There was nothing unusual in that.
Great. Thank you.
Thanks. Next question, please.
The next question comes from Larry Greenberg from Janney Montgomery Scott. Please go ahead.
Thanks, and good morning. I'm not sure there's anything else to ask specifically on your quarter. I'm just curious, with social inflation being the big topic in the underwriting world, wondering if you guys could either take your intermediary experiences or go back to your underwriting days and maybe just give your perspective on what's going out there, and whether there's any analogies you would draw to past periods where the industry might have experienced some of these similar challenges.
Well, John has forgotten more about underwriting than I ever knew. John, you want to take that?
Look, it's apparent to me that the underwriting community is struggling to figure out what the trend line is, right? Much like I think I said pricing, it's not one market, it's a collection of markets. I think the challenge is it's not one trend line, right? As I said earlier, there's some pretty big, very material casualty losses in the market. Then you're seeing where you have more data and more frequency in commercial auto, a clear trend that's emerged over the course of the last several years. In the case of auto, it's got a shorter duration to it, as you know, than other liability lines. It's quite, again, clear that they're trying to sort that out. Again, our responsibility is serving our client and trying to come up with the best outcome for them in this market.
I would expect a bit more stress in the near term in the excess liability market in particular. There are some areas where you can point to meaningful exposure changes, right? In the case of D&O, maybe, for example, there's been a material increase in the frequency of securities claims that are in the market, right? That obviously, in the aggregate, needs to be priced for. There are some segments where you're seeing clear exposure changes, in other areas of risk, you're just seeing higher settlements and judgments for a lot of reasons.
Yeah. We were talking recently about workers' comp. In time, for one reason or another, and you can have a lot of opinions about what's underneath it, in times of full employment, workers' comp claims tend to go down, and in times of economic stress, they tend to go up. Workers' comp claims are in decent shape now, that doesn't mean it's a permanent trend line, and that may reverse with economic stress.
Yeah. I would also point out, Dan, that after the financial crisis, we had a period where there wasn't real loss cost inflation, right? Claim trends were quite stable. In fact, in some lines of business, there was deflation, which isn't what we normally observe over a long period of time. Clearly, we've come out of that environment.
Anything else, Larry?
No, that's good. Thank you.
Thanks.
I would now like to turn the call back over to Dan Glaser, President and CEO of Marsh & McLennan Companies, for any closing remarks.
Sure. Thanks, Calina. I just want to thank everybody for joining us on the call this morning. I want to express my gratitude to our 75,000 colleagues for the commitment and hard work that they show us all the time, as well as to our clients for their support. Thank you all very much, and we'll speak to you next quarter.
That will conclude today's call. Thank you for your participation. You may now disconnect.