U.S. Markets closed

Edited Transcript of GCAP earnings conference call or presentation 24-Oct-19 8:30pm GMT

Q3 2019 GAIN Capital Holdings Inc Earnings Call

Bedminster Nov 6, 2019 (Thomson StreetEvents) -- Edited Transcript of GAIN Capital Holdings Inc earnings conference call or presentation Thursday, October 24, 2019 at 8:30:00pm GMT

TEXT version of Transcript

================================================================================

Corporate Participants

================================================================================

* Glenn H. Stevens

GAIN Capital Holdings, Inc. - President, CEO & Director

* Nigel Rose

GAIN Capital Holdings, Inc. - CFO

* Nicole Briguet

Edelman - Account Supervisor, Financial Communications

================================================================================

Conference Call Participants

================================================================================

* Daniel Thomas Fannon

Jefferies LLC, Research Division - Senior Equity Research Analyst

* Kenneth Brooks Worthington

JP Morgan Chase & Co, Research Division - MD

* Kyle Kenneth Voigt

Keefe, Bruyette, & Woods, Inc., Research Division - Associate

* Rajiv Sharma

B. Riley FBR, Inc., Research Division - Analyst

* Richard Henry Repetto

Sandler O'Neill + Partners, L.P., Research Division - Principal of Equity Research

================================================================================

Presentation

--------------------------------------------------------------------------------

Operator [1]

--------------------------------------------------------------------------------

Good day, everyone, and welcome to the GAIN Capital Third Quarter 2019 Results Conference Call and Webcast. Today's call is being recorded.

At this time, I would like to hand the conference over to GAIN Investor Relations representative, Nicole Briguet. Please go ahead.

--------------------------------------------------------------------------------

Nicole Briguet, Edelman - Account Supervisor, Financial Communications [2]

--------------------------------------------------------------------------------

Thank you, operator. Good afternoon, and thank you to everyone for joining us for our third quarter 2019 earnings call. Speaking today will be GAIN Capital's CEO, Glenn Stevens; and CFO, Nigel Rose.

Today's commentary will be accompanied by our earnings slide deck, which can be accessed via webcast on our IR website now or at a later time.

Following their remarks, we will open the call to questions.

During this call, we may make forward-looking statements to assist you in understanding our expectations for future performance.

These statements are subject to a number of risks that could cause actual events and results to differ materially.

I refer you to the company's Investor Relations website to access the press release and the filings with the SEC for a discussion of those risks.

In addition, statements during this call including statements related to market conditions, changes in regulations, operating performance and financial performance, are based on management's views as of today, and it is anticipated that future developments may cause these views to change.

Please consider the information presented in this light.

The company may, at some point, elect to update the forward-looking statements today, but specifically disclaims any obligation to do so.

I'd now like to turn the call over to Glenn.

--------------------------------------------------------------------------------

Glenn H. Stevens, GAIN Capital Holdings, Inc. - President, CEO & Director [3]

--------------------------------------------------------------------------------

Thanks, Nicole, and thanks to everyone for joining us today.

During the third quarter, we saw continued positive signs of client engagement. In our retail segment, we continued to deliver new account growth, which nearly doubled over last year and grew 32% sequentially, positioning us well for the return of normal market conditions.

Pockets of volatility in U.S. equities supported growth in our Futures business, which saw a 24% increase in average daily contracts in the quarter compared to last year. Turning to our financial results. Q3 net revenue came in at $66.7 million with an adjusted EBITDA of $6 million.

Turning to Slide 4. In Q3, our marketing investments delivered the fifth consecutive quarter of new account growth. In addition, trailing 3 months active accounts improved 5% over last quarter. We are encouraged by these results and believe in our continued success driving organic growth.

On the next slide, digging into our active customer base. Our continued success in acquiring new customers has been translating into consistent growth in our active customer numbers.

As you can see on the chart, our trailing 3-month actives increased for a third consecutive quarter, putting them 10% above the level last in Q4 of '18. Notably, actives were above Q2 levels of '18, which was the last full quarter prior to the implementation of ESMA regulations.

We remain well placed for when volatility returns. A good short-term example being the recent oil price move tied to the missile strike in Saudi Arabia in September, where our platform experienced a fivefold spike in crude oil volumes. Examples like this prove that we have a sizable stored client potential, which will materialize when volatility returns.

In addition to the new and active account growth we're seeing, the 4 other key metrics we use to evaluate our marketing performance remain on track, and they are:

Our cost per new account. After 5 quarters of increased spending levels, we continue to track below our targets.

Our breakeven point. The timeline to recover the marketing investment from each cohort remains on track with Q3 results in line with our expectations.

Our internal rate of return. Our more recent IRRs on customer acquisition costs continue to exceed our expectations and validate our marketing spend as a strong use of our capital.

And finally, our long tail of revenue. While a healthy percentage of our revenue continues to come from long tenured clients, as our new accounts become active, the mix of client transaction revenue has begun shifting to our newer customer base. We continue to expect that the newly acquired customers will deliver revenue well beyond our ROI benchmark, which is based on a 3-year lifetime value.

In addition, we continue to analyze ways in which we can optimize our spend in local markets to increase yield and/or reduce our spend with limited or no negative impact. Consequently, given the prevailing usually -- unusually low volatility environment, we are targeting 2019 annual marketing spend to come in closer to $41 million or up about 16% over last year.

And in the past, we have spoken about the AI-driven hedging model that we have developed and deployed. Thanks to our focus on leveraging data to drive more effective pricing and improved risk management, our new hedging model continues to outperform our past approach to hedging activities. Our stated objective for this effort was to reduce variability of revenue capture. Two commonly used measures for variability and risk management are standard deviation and the Sharpe ratio. Our standard deviation of daily revenue is now more than 25% lower under the new model, while our Sharpe ratio is almost 50% higher. So in both cases, we're seeing marked improvements.

Fortunately, improved variation metrics are not expected to come at the expense of our long-term revenue capture, which will continue to trend in line with our previous guidance of $106 RPM.

We are pleased with our progress to date and will continue to apply this model to a broader range of products in our portfolio. We originally applied our new model to FX only, and we expect metals and indices will go live this quarter.

Looking forward, our long-term strategic priorities to accelerate growth remain intact. The 4 pillars of our growth are rooted in our data-driven approach to customization. A few notable examples. With our marketing investment, we're leveraging detailed cohort payback and internal rate of return analyses to enhance and customize our spend at the regional and product offering levels. On innovating the trading experience, we're using data to get a better understanding of how our customers trade and what their preferences are to enhance our platforms to better suit their requirements on a customized basis. By personalizing our trading experiences based on regional, product and customer segment preferences, we can drive higher engagement and activity, thereby extending lifetimes and lifetime value.

I mentioned a great example of this earlier with the recent price movement in oil. In August, we introduced a new spot crude product to complement our existing Futures offering and provide our customers with a product they more easily recognize. This affords us the opportunity to take greater advantage of the spike in activity, with that new product accounting for almost half of the volume during that short elevated period. Data is also helping us become more proactive in identifying potential premium clients much earlier in their journey with us and helping us understand how to better engage with them.

As I reflect on the past 9 months and look forward, I'm encouraged about our position to deliver long-term value. We acknowledge that we have witnessed a period of unusually low volatility that has had an adverse impact on our ability to deliver short-term financial results.

With that said, we have not watched from the sidelines, but rather have been proactive in evolving our business model and our marketing efforts that has grown our customer base, which is critical to the long-term success of this business.

Ultimately, our broadening base of customers will drive event-driven volume increases, enhancing the benefit from our improving operating leverage as we reduce our expense base even further. We have repeatedly delivered expenses at or below the lower range of our guidance and expect to continue to do so. Our ability to better capture revenue opportunities, coupled with expense management, will ultimately allow us to maximize profitability.

With that, I will turn it over to Nigel for a deeper review of our third quarter results. Nigel?

--------------------------------------------------------------------------------

Nigel Rose, GAIN Capital Holdings, Inc. - CFO [4]

--------------------------------------------------------------------------------

Thanks, Glenn. During the third quarter, net revenue decreased 30% year-over-year to $66.7 million as compared to $95.5 million last year. GAAP net loss was $2.1 million, resulting in GAAP loss per share of $0.06 as compared to net income of $10 million and GAAP earnings per share of $0.22 in the prior year. Adjusted loss was 2.1 -- $2.8 million and adjusted loss per share was $0.07 as compared to adjusted net income of $13.8 million and an adjusted EPS of $0.31 in Q3 2018. Adjusted EBITDA was $6 million compared to $30.5 million in the prior year.

In our retail segment, market conditions saw a quarterly ADV decrease of 10% year-over-year to $7 billion. RPM of $114 for the quarter was slightly above our long-term expectations, but below the standout RPM of $164 seen in Q3 2018. The combined impact of ADV and RPM saw total retail revenue of $57 million, a year-over-year decrease of 34% with the RPM differential accounting for almost 90% of that reduction.

Marketing for our retail segment was up 36% for the first 9 months of 2019 as compared to the prior year period as we remained committed to our organic growth strategy. Retail overheads for Q3 and the year-to-date were both down 6% over the prior year, reflecting our continued focus on our operational efficiency strategy.

Turning to the Futures business. Revenues were up 9% at $10.6 million for the quarter as compared to $9.7 million last year. While Futures average daily contracts increased 24% to 31,895 during Q3, revenue per contract decreased to $4.59 due to shift in product mix, but remains in line with the year-to-date and trailing 12-month averages.

Overheads for the Futures segment year-to-date were 4% lower over the prior year period, and profit margin for the Futures business improved to 16% in Q3 2019 and up slightly to 15% on a year-to-date basis compared to prior year.

I'd like to take a moment to provide an update on our overhead guidance. On our Q3 earnings call, as a result of continued optimization of our cost structure outside of our growth initiatives, we guided to overheads this year coming in near $180 million compared to the original guidance provided during Q4 '18 earnings of $190 million to $200 million.

With 1 quarter left in the financial year, our overheads stand at $132 million, equating to an annualized $176 million. The further reduction below the previous $180 million-plus guidance has been driven in part by a reduction in the variable component arising from financial performance this quarter. At this stage, our 2020 guidance remains within the range previously provided, namely $170 million to $180 million.

We remain committed to pursuing opportunities to further streamline our overhead expenses, which we believe will ultimately enable us to create stronger operating leverage as well as allow us to be profitable should the unusually weak market conditions continue.

As a reminder in Q1, we decided to amend the presentation of our liquidity to directly align with the cash and cash equivalents' figure shown on the balance sheet. We made this change as we believe it's preferable to reference the cash reported on our balance sheet and how that has moved period-over-period as set out in the appendix to this presentation. We continue to balance the uses and needs for our cash in the order the diagram on this slide shows. The primary priority being to ensure GAIN has the liquidity it requires to operate efficiently and effectively.

At the end of the quarter, GAIN had total cash and cash equivalents of $200.7 million. Broker receivables continue to remain at heightened levels, being $107.7 million at the end of Q3 2019, almost double the prior year's $57.5 million.

This reflects an increase in the size of our hedged positions caused by the growth in open client positions, which is a good lead indicator for future revenue. Secondly, we remain committed to actively returning capital to shareholders through dividend payments. As such, our quarterly dividend of $0.06 will be paid on the 17th of December. This represents the 32nd consecutive quarter GAIN has paid a dividend, a period stretching back to 2011.

In terms of the next priority, buybacks, GAIN did not repurchase shares during the quarter as we believe it remains more prudent for now to retain capital during this prolonged period of market -- low market volatility. In addition, we are conscious of our $60 million convertible loan falling due next April that we intend to repay, which will also reduce the interest expense impact that has on our profits.

However, we'd still remain committed to actively returning capital to shareholders with the buyback being a key component of that. A good example being we recently bought back a portion of the 2020 converts. In terms of share buybacks, we have approximately $41 million authorized and remaining for additional repurchases as of September 30.

With that, I will now turn it to the operator for questions.

================================================================================

Questions and Answers

--------------------------------------------------------------------------------

Operator [1]

--------------------------------------------------------------------------------

(Operator Instructions) And the first question comes from Dan Fannon with Jefferies & Company.

--------------------------------------------------------------------------------

Daniel Thomas Fannon, Jefferies LLC, Research Division - Senior Equity Research Analyst [2]

--------------------------------------------------------------------------------

I guess my first question is on capital, the no buybacks in the quarter. Just curious as to why you're just looking at the share price action. And then ultimately, just thinking about M&A in the construct of a low-vol environment. You've talked about how that puts pressure on some of the more -- the smaller competitors with less scale. So can you talk about the environment today for potential M&A as well?

--------------------------------------------------------------------------------

Glenn H. Stevens, GAIN Capital Holdings, Inc. - President, CEO & Director [3]

--------------------------------------------------------------------------------

Sure, Dan. On the -- they actually tie in a little bit together in that, number one, because we don't feel like we have an option using our equity at these prices for M&A that leaves cash -- or seller financing or other creative measures -- but particularly cash as our best option for opportunities that arise in this low-vol environment. So that dovetails also with our decision to temporarily scale back some of our -- some of the buying back our own stock, which we've been, number one, pretty consistent with participating in, and it's not off of our slate of options going forward by any measure.

But there -- 2 of the items. Number one, some of the M&A that you mentioned, trying to be prudent and timely for that. And the second part of it is we are planning to pay off the first tranche of our debt in April of '20, which is about $50 million or actually a little higher than that, about $60 million to pay off that full amount. We bought a small portion back of it, but essentially making sure that that's kind of earmarked and geared to be paid off. So we're being mindful of that as well in this environment where it's more difficult to generate a lot of excess cash. So on both sides of those, we made the decision over this quarter to do a little bit of wait-and-see.

So I guess the combination of the 2. Number one, being prudent and having that available without changing any of our operations or strategic levers to pull. And number two, from an M&A perspective, you're right that the environment is interesting for us, but we -- I don't believe we have, at present value, stock is one of our levels -- levers to use.

--------------------------------------------------------------------------------

Daniel Thomas Fannon, Jefferies LLC, Research Division - Senior Equity Research Analyst [4]

--------------------------------------------------------------------------------

Got it. And then just with regards to expenses, I understand the marketing and the outlook for that in the fourth quarter. I guess just surprised in the third quarter that the G&A number went up sequentially as much as it did, and just if there's anything behind that or kind of thinking about G&A going forward and then in the context of the overall overhead cost and the other kind of framework you gave us, Nigel.

--------------------------------------------------------------------------------

Nigel Rose, GAIN Capital Holdings, Inc. - CFO [5]

--------------------------------------------------------------------------------

Yes. No, it's good question, Dan. You're right. It did step up in Q3. In part, there were some one-off costs in there, which we, by its nature, won't repeat. And in part, there was a bit of a step-up within a variable component in relation to bank fees and the fees we pay our payment providers for processing customer deposits and withdrawals, which we saw as a positive and as much as we're seeing a little bit more engagement from those guys now in terms of their funding and trading to some degree.

--------------------------------------------------------------------------------

Operator [6]

--------------------------------------------------------------------------------

And the next question comes from Rich Repetto with Sandler O'Neill.

--------------------------------------------------------------------------------

Richard Henry Repetto, Sandler O'Neill + Partners, L.P., Research Division - Principal of Equity Research [7]

--------------------------------------------------------------------------------

I guess my question is first on Slide 7, when you show the hedging, the AI-driven hedging model. And I get you're looking at the standard deviation of daily revenue. But when I saw -- just take a look at your RPM over the same period, I mean it certainly looks like that's very volatile when you had a 50 and a 164. So I guess I'm not up on my statistics here. But can you explain, sort of, the difference between how we get to the numbers for the quarterly RPM when everything in your chart shows we're reducing volatility?

--------------------------------------------------------------------------------

Glenn H. Stevens, GAIN Capital Holdings, Inc. - President, CEO & Director [8]

--------------------------------------------------------------------------------

Totally fair question. Part of it is perspective. And so what I mean by that is on -- and they're slightly 2 different approaches. On the one hand, the standard deviation that is reported daily was -- is ultimately designed to drive a narrower range over quarters and over a longer period. So in not so much in a vacuum but from a daily measure, we have seen a relatively consistent march into a lower standard deviation of our P&L day-to-day. Now we don't report a daily P&L but, of course, we track our own daily P&L; and ultimately by driving a narrower range of outputs there, that will stack up into a narrower range of variability from, in this case, quarter-to-quarter. That's going to be, kind of, over time.

The reality is that the older model, go back Q1 of '18, was -- we actually have a wider range by -- for daily standard deviation of P&L, and it actually did drive a higher -- sorry, lower Sharpe ratio. So you want lower standard dev, higher Sharpe.

Now to your point, when you look back for the last 2 quarters, for example, and look at the 164 and the 50 and you add the 2 of them together, it goes right smack again to our average of 105. So yes, on 2 quarters with an up and a down, you'd say, oh, gee, that's moving around a lot. When you take the lens back a little further and look back and smooth it out into multiple quarters, it goes right smack into the magnet of 105 that we've been tracking to for years, frankly.

So on that one, over those 2 quarters, no, you're right that this particular model didn't soften that. Reality, though, is, believe it or not, those spikes actually would have been more exaggerated, just given market conditions and such, that it would have actually been wider than 50 and 164. It's not an improvement over the previous 3, 5 quarters, but actually was an improvement over -- because we run simulation testing all the time, and was an improvement over the old model that way.

So again, on the longer -- if we had a parallel universe of looking the old model, new model, this new model does continue to actually raise the Sharpe and lower the standard deviation. Those 2 quarters that you pointed out, probably fair. But when you look back further, and you -- and we go back 1, 3, 5, 7 years, new model testing complete -- outperformed in every case.

--------------------------------------------------------------------------------

Richard Henry Repetto, Sandler O'Neill + Partners, L.P., Research Division - Principal of Equity Research [9]

--------------------------------------------------------------------------------

Okay. That helps, Glenn. Then the next question, Nigel, just on the tax rate. Can you tell -- you had -- I think you had given this guidance for the 3 quarters to date, and it's sort of varied, was that because of the loss, I guess?

--------------------------------------------------------------------------------

Nigel Rose, GAIN Capital Holdings, Inc. - CFO [10]

--------------------------------------------------------------------------------

It -- yes. I mean it's a function of the level of losses, but also where those losses arise, and then the impact of items that are deductible for P&L purposes but are not deductible for tax purposes. And those can move the rate around a little bit because I -- you would think 19% in the U.K., 21% in the U.S., you'd be around 20% ordinarily. On Slide 23, year-to-date on adjusted basis, we're at the 16%, which I think we mentioned on the September metrics. So the gap between the sort of 20% and the 16% would be due to some of those nondeductible items reducing the tax credit.

--------------------------------------------------------------------------------

Richard Henry Repetto, Sandler O'Neill + Partners, L.P., Research Division - Principal of Equity Research [11]

--------------------------------------------------------------------------------

Got it. Okay. And I guess sort of off-the-wall question, Glenn. You certainly have heard about 0 commissions in the equity market. And I'm just trying to see whether any -- I know your model is completely different. You're not in a district agency model-like, but has that even come up with clients in regards to the lowering of commissions, I guess, in a different asset class?

--------------------------------------------------------------------------------

Glenn H. Stevens, GAIN Capital Holdings, Inc. - President, CEO & Director [12]

--------------------------------------------------------------------------------

It's an interesting point, Richard. Of course, we've been watching that whole development unfold so quickly the way it did with such large players for a piece of the market, particularly in the U.S. One little side benefit that it's actually had is that there was a certain sense of -- we can either call it mistrust or lack of understanding or a discomfort where actually customers not paying commissions in our business kind of didn't sit well with some customers.

You have new customers looking into our market, trying to figure out how to trade this market, and it's almost like I don't get it, that seems too good to be true. I totally get it if you charge me $5 or $10 a trade, then it would make sense. What do you mean there's no commission? And in trying to walk somebody through the whole concept of being able to have a bid offer spread and extract our de facto commission on each trade by being the market maker, it's hard to explain them saying it's $7.99.

So in one -- in some ways, now we have others joining our challenge to explain to customers to say, well, yes, we're able to let you trade with us without commission. And they say, okay, I get it; but why? And then they have to explain security lending revenue and financing revenue and all the kind of stuff, not great and not easy. So in one -- in some ways, it actually adds credibility to our models like, oh, I see, I see, how a broker can be commission-free and still be completely above board. So that's kind of an odd but beneficial understanding for us as equity traders look at other markets.

The other part of it, though, is that outside the U.S., we do have commission -- some commission-based products where people are accustomed to paying commissions to trade non-FX products. And in those cases, we have a situation where the spreads are so tight that they'll pay that, and we have yet to see any of the providers in that CFD business feel any pressure to adjust the models. But it is something we would see -- we would have to consider. And my guess is that the pricing would just be adjusted.

So when you have these razor thin bid offer spreads that -- in these products that have a low commission, you probably end up with an adjustment that way because it's a little bit of a combination of is it in the spread or is it in the commission, until it ends up getting adjusted. But in general, it hasn't created any negative connotation. I just wanted to highlight that one kind of positive one because our relationship managers have provided some feedback to say, hey, people got this question and kind of said like, oh, gee, now I get it.

--------------------------------------------------------------------------------

Richard Henry Repetto, Sandler O'Neill + Partners, L.P., Research Division - Principal of Equity Research [13]

--------------------------------------------------------------------------------

Got it. Got it. And Glenn, I'm going to leave it to someone else to ask how October is doing to date. So...

--------------------------------------------------------------------------------

Operator [14]

--------------------------------------------------------------------------------

And our next question comes from Kyle Voigt with KBW.

--------------------------------------------------------------------------------

Kyle Kenneth Voigt, Keefe, Bruyette, & Woods, Inc., Research Division - Associate [15]

--------------------------------------------------------------------------------

I'm not going to ask a question on October, but just a question on the new direct accounts that continue to grow kind of in that slide you provided there. I guess when we see a number above, I think, it's above 40,000 on this new direct accounts acquired in the quarter. But we only see your trailing 3-month active direct accounts increased by 3,000 in the quarter. Because I'm trying to figure out what does that really mean. I guess it's one of 2 things. Are these clients just opening accounts and just not trading yet? Or are you experiencing higher attrition rates than you have historically?

--------------------------------------------------------------------------------

Nigel Rose, GAIN Capital Holdings, Inc. - CFO [16]

--------------------------------------------------------------------------------

Hi, Kyle. Yes, I mean, I'm sure Glenn can weigh in, but the -- it's a good question. I mean it does take time for new accounts, once they've opened accounts, to get used to the platform, going on fund and then feel that they are ready to trade and then put their own money into retrading the markets. So that can be a slight delay.

--------------------------------------------------------------------------------

Glenn H. Stevens, GAIN Capital Holdings, Inc. - President, CEO & Director [17]

--------------------------------------------------------------------------------

And I would mention that that kind of delay, which we do monitor very closely, in quieter periods almost automatically stretches. So when it's really busy, somebody opens funds and trades in a very short amount of time. And when it's low volatility or not as much of an urgency effect, the time between submitted app for trade, which, by the way, in between there means approval and funding and traded, extends out.

Now we can accelerate by our own technology and efforts and investments the ability for us to approve a submitted app, but that doesn't help the time line as much because we can't extract funds from somebody. They have to opt to fund obviously voluntarily. So without that now factor in the markets that naturally extends itself, which we've seen in this environment. So that's kind of one factor.

The other factor, I think, I would add also is that some of the customers we're bringing into our market as we widen our audience, and we've expanded our marketing, does bring in slightly less intended or urgent customers who are seeking us. We know because by natural selection when we cast a wider net, we have people who say, yes, that is pretty interesting. I'll open an account. But what happens is, we -- they almost end up in our funnel of conversion efforts where we have outbound. So someone onboards, put in an app -- and this is nothing new for anybody who has shopped online for a vacation or a piece of clothing or whatever, where you get an e-mail that says, hey, we saw you're interested, what can we do to sweeten the deal or how can we get you to convert to a live customer?

And so again, this concept of, number one, there not being a market urgency; and number 2, going to a wider audience. There's a little bit more effort to get them to that traded level because I will say that the funding percentages have held up pretty well, which is encouraging. Because on one hand, it's been a while, but we no longer tout unique impressions, right? Remember that term everybody was using in online businesses, unique impressions, like who cares, people are coming to your website, what are you doing to convert and then monetize them?

So we moved on from there. And then it was submitted apps. Yes, but if the quality of these submitted apps degrades, then that doesn't help you convert. We haven't seen that. That's actually holding up very well. And then some -- the percentages of conversion from submitted apps to approved apps has held well. Because again if your quality was low, people aren't properly well-versed, they're not of age, they don't have sufficient funds, whatever things that would disqualify them in various jurisdictions has held up well. And then the funded percentages has held up well.

The part that's the challenge for us in these environments when it's quiet is getting converted to trade. So it doesn't surprise me that that yield is lower. But what's really more important is that your hurdle to engage them, and that's why we use the example of oil.

We didn't have a double or a triple; we had 5x spike in oil. Now granted, oil is a small product for us, and it was 2 days, and it didn't have a demonstrative P&L impact, obviously. However, it did show me that with a broader base of opened and funded and ready to trade clients, when something happened that was compelling, and that happened on a Saturday, and by Sunday evening, we saw this huge spike that continued through Tuesday that show a bigger base of clients engaged.

And by the way, we looked at the numbers, and it wasn't just existing customers that were with us for more than a year. Nearly half of the volume came from customers inside of the year. So even new people who hadn't necessarily engaged a lot and said, oh, shoot, I have an opinion. I need to do something. So I think that was really helpful.

And then when you add on top of that kind of us trying to innovate a new product, and that's why we highlighted the idea of saying, hey, here's a spot oil product that doesn't roll over like a future. It's just really easy to trade. It trades just like a currency. You're bullish oil you're bearish oil; you don't have to worry about carry, financing, end of contract life, or anything like that. We can see that the response is really good.

So I guess what I'm -- I use it as a microcosm. But the reality is, as long as that potential value is there, which we're seeing by building all these funded accounts -- and that's why we count the funded accounts. They're not open applications, they're not submitted applications. They're actually accounts that have been completely approved and are ready to go.

--------------------------------------------------------------------------------

Kyle Kenneth Voigt, Keefe, Bruyette, & Woods, Inc., Research Division - Associate [18]

--------------------------------------------------------------------------------

Okay. And then I guess, I think part of that question, I don't know if you directly answered or not, but maybe I'll more directly ask it is have your attrition rates increased? I guess when you look back over the past 18 months or 24 months, has there been an uptick in attrition rates?

--------------------------------------------------------------------------------

Glenn H. Stevens, GAIN Capital Holdings, Inc. - President, CEO & Director [19]

--------------------------------------------------------------------------------

It's been -- uptick in attrition. No, not necessarily an up -- so like the rate of churn hasn't materially changed from what we've seen. You often see it at lower levels of customer attraction.

--------------------------------------------------------------------------------

Kyle Kenneth Voigt, Keefe, Bruyette, & Woods, Inc., Research Division - Associate [20]

--------------------------------------------------------------------------------

Okay. And then just on the marketing. I think you've -- maybe you didn't put it in this slide. But in last quarter, I think, you talked about an IRR, very high IRR on the marketing spend and you highlighted a good use of capital right now. It is continuing to invest in marketing. I guess just -- is that -- what assumption is going into that in terms of you just talked about these -- these are different types of clients maybe that you're kind of casting a wider net. Is it [possible] that these clients may be a lot less active than your current clients, maybe less of them just even convert from funded accounts to actually trading or active accounts? Just wanted to know the assumption are in that.

--------------------------------------------------------------------------------

Glenn H. Stevens, GAIN Capital Holdings, Inc. - President, CEO & Director [21]

--------------------------------------------------------------------------------

No. It's a good question, Kyle. And the way we measure it is a couple of fold. One of them is just time. And so we're just coming up on our first cohort of a year's worth. So if in Q4 of last year, someone is now coming up on their first full year with us. And if we are looking at a 3-year -- we generally consider lifetime value of our clients to be mostly captured within 3 years. Now the reality is we have lots of clients who trade with us more than 5 years. And there's this long tail of revenues is almost -- almost 50% of revenues come from what we consider loyal clients with a tenure of more than 3 years.

But for the purpose of this marketing investment, we use a 3-year value as their -- as the total return aspect to look to deduce the IRR over that time, right? So we look at the amount of money. The amount of spend, the amount of return in a 3-year period yields the IRR. The good news is that that's held up, continues to hold up, above our expectations and pretty well.

So in terms of the degradation of the quality of the client, no, we're not seeing a market -- marked situation there. However, to be fair, to truly measure it, we probably have to look out another couple of quarters to say, okay, when we add that first quarter cohort that came in, let's call it a Q4 cohort then it was a Q1 '19 cohort, Q2 -- to Q2, it's hard to make a complete determination that early in the cycle, too, when you're -- and the reason being, I think I have a better answer for you that if there were a plethora of trading opportunities, and the returns weren't there, then I'd have to say to you, yes, you know what, hmm, because they should have been more active, they should have traded more, then we should have generated more revenue we call CTR, client transaction revenue. Thank you. I forgot that one.

But client transaction revenue, it should have been there. But when you -- you don't have that plethora of opportunities. And instead, you have a dearth of opportunities in long term, low vol, then you don't want to be quick to judgment to say, hey, you're a week -- you're a quarter in or 2 quarters in. Well, first of all, it's 2 quarters over 12. And second of all, you're in this pretty poor environment to be able to make a decent judgment.

So I'd say that based on observation empirical evidence, no. And number two, based on kind of subjective, too, to say, well, you haven't really even had a decent test that way. So despite that, they're still getting some pretty solid returns in terms of IRR on the investment, and that's why we actually mentioned it again in this deck to say we still consider it a very -- a rewarding use of capital for us.

--------------------------------------------------------------------------------

Kyle Kenneth Voigt, Keefe, Bruyette, & Woods, Inc., Research Division - Associate [22]

--------------------------------------------------------------------------------

Okay. I have one more question, and then I'll hop back in the queue. But this one's maybe a little more unusual, but when I'm thinking about your business and maybe potentially other adjacent areas where kind of like the regulatory status scale, trade, technology, user interface could be useful, one that comes in mind is online and mobile sports betting. And another broker in the space, IBKR, launched a simulated sports betting application recently.

I guess my question is, would there be any desire for GAIN to build out something to address that kind of growing sports betting market in the U.S. directly or even licensing some of your technology or partnering with someone in that space? Or is that just too far of a leap?

--------------------------------------------------------------------------------

Glenn H. Stevens, GAIN Capital Holdings, Inc. - President, CEO & Director [23]

--------------------------------------------------------------------------------

It's a really interesting question. And we have spent a fair amount of mind share internally reviewing it, looking at it with partners. It's amazing, frankly, to see how much of a magnet it's drawn. I don't think you can listen to the radio or watch television -- I don't think it's inundate online yet, frankly -- but certainly on TV and radio in this tristate area, you can't go 11 seconds without somebody hitting you up with a client customer bonus.

And in our business, to be able to track a lot of these marketing activities, we're very accustomed with it, and it's amazing how similar they are. They're deploying, frankly, financial services models to say, look, here's how you refer a customer or here's a signing bonus or here's a way to make a trade de facto but not lose any money on it. All these kinds of things are amazing. And I'm not so sure if FINRA or the CFTC would be all too pleased, but neither of those entities regulate sports gambling.

And so the short answer to your question is, yes, we very much have considered it. And it was interesting to see IBKR dip a toe, if you will. Also to note some of our U.K.-based brethren, in the past, actually had pretty sizable sports betting operations. And what's interesting, you spent -- up until about 5 years ago, 7 years ago, it very much was a part of their business, and then exited it because there was this kind of arguable tarnish that said, no, we're for investors, and so we're not going to get into sports betting.

And yet, you see very reputable firms now kind of circling back and say, hey, if that's what you want to speculate on, with all these tools and all this technology, it's much more than just a guy on a street corner taking a bet for you and giving you a slip of paper.

Then I guess, my answer to you is, I don't believe it's outside of the realm of possibility. We do have a lot of technology and expertise in this space. However, it does appear that some of the private equity-backed shops and some of the other on transports from other locales have really flooded this market. It'd be interesting to see if it's able to support all of them because there's got to be 15 -- right off the bat, 15 pretty big providers for just, at least for now, Pennsylvania, New York and New Jersey, well, really just Pennsylvania and New Jersey market.

So -- but look, to be fair, particularly in the U.S., it's highly likely this becomes a multistate kind of like E-ZPass, pretty quickly. So it wouldn't be a terribly hard pivot for us, and it's something that if it's done in a way that we think we can add value and differentiate, I don't think it should be off our possibility list.

--------------------------------------------------------------------------------

Operator [24]

--------------------------------------------------------------------------------

(Operator Instructions) And the next question comes from Ken Worthington with JPMorgan.

--------------------------------------------------------------------------------

Kenneth Brooks Worthington, JP Morgan Chase & Co, Research Division - MD [25]

--------------------------------------------------------------------------------

We have seen, I guess, losses 3 out of 4 quarters. There's clearly a lot of pressure on the business. How does the consolidation environment look right now? So as you look to your, I don't know, smaller competitors or the competitors in parts of Asia and Europe, I assume the pressure is fairly intense on them as well. Are those competitors closing shop? Does it make sense to acquire that -- those -- that customer base? So what is your appetite for transactions right now?

--------------------------------------------------------------------------------

Glenn H. Stevens, GAIN Capital Holdings, Inc. - President, CEO & Director [26]

--------------------------------------------------------------------------------

The appetite and the ability remain because we have a track record of, I would argue, successfully assimilating acquisitions over the last decade, it does give us comfort and faith that when and if they make sense, we'd be able to move forward and do something about it: number one, maintaining a fair amount of cash dry powder; number two, being operationally willing and able to do those deals. Because by the way, sometimes people are financially able but operationally they get mired. And so that's handy.

And we certainly have pockets of opportunities globally where they would make sense, too. So regions like Southeast Asia and Middle East and Latin America where -- even certain parts of Europe, where we're not particularly strong yet. I would argue that it's a step function ability to get into those markets.

The good news is that with a prolonged period, as you mentioned, of challenges in this environment, it does make for a more reasonable value discussion when you have these discussions with people. However, in some ways, when you buy another business, we're pretty good at being able to price the assets of another business. But when someone wants to sell their intrinsic, what they consider their intrinsic value in technology or market positioning or what have you, that's a harder discussion.

Because for us, we kind of feel like it's a 2 pressure. One pressure is prevailing market environment at -- in the present. The other pressure is the regulators who actually create a benefit for larger, more stable, more solvent providers like ourselves. And so in some ways, the marginal players who are deploying very aggressive tactics or very thinly capitalized or not really providing the proper infrastructure for an on-the-boots organization.

I'll give you an example. You even have outside policemen like Google, determining that if you don't have a license in a market, you can't be found in a Google search in that market. So let's use Canada, for example. Unless you are properly licensed there, only recently Google shut down advertisers there who decided to flout the local registration, and would try to attract clients like that.

Well, the good news is, for us, is that we become part of a very small group that can provide these services there because we are properly registered and licensed. And all those people we were competing with who didn't have any audits that come from IIROC and capital requirements and staff and such were competing with us at a very low cost basis, now they can't. Just because the regulator does what they can, but it really helps when Google says, well, no, we're just going to block your content.

So that's good examples of we don't need to buy that shop in that business there because we're going to get that business going forward. I'm not saying it's a huge market opportunity, but what I'm saying is it's an example, clearly, of other outside pressures forcing some of the marginal players out. So I think that we do -- are very open and active in discussions where it makes sense strategically. But there's also a bit of a positive coming our way.

And I guess I'll add, too, being well-capitalized does help kind of short term challenges when it comes to making money, losing money on a -- even on a cash basis, making money; but on a noncash basis, a net income basis when you count D&A and others losing money. But it doesn't mean that we are also not finding ways to lower our breakeven. We continue to do that. Actually, if you compare our kind of present financials with ones even a year or 2 ago, we're -- by managing our cost effectively. And you'll notice we continue to come in in the lower end of our -- every time we give our cost guidance, we come in in the lower end of it. That's not a desire that's going to go away from us.

--------------------------------------------------------------------------------

Nigel Rose, GAIN Capital Holdings, Inc. - CFO [27]

--------------------------------------------------------------------------------

Yes. And just to add to what Glenn said. We focused on overheads, but if you look at our income statement now you'll see D&A has come down because we've been conscious about our CapEx spend and reduced that over the last couple of years, so that's flowing through into the D&A number. And as we mentioned earlier, repurchasing the converts, that then reduces the interest expense. So we're looking at every line we can in the income statement to see how do we get those to a level that in conditions where the ADV is $7 billion, we could be fairly confident we'll have a positive net income.

--------------------------------------------------------------------------------

Glenn H. Stevens, GAIN Capital Holdings, Inc. - President, CEO & Director [28]

--------------------------------------------------------------------------------

With the idea being that if the storm -- you can't determine how long challenging conditions prevail. But we want to be able to be in a better situation so that when the weaker providers continue to suffer, I'd be able to do a deal with them or they can't compete with you anymore. So that's why it's got a benefit that way, too.

--------------------------------------------------------------------------------

Operator [29]

--------------------------------------------------------------------------------

And the next question comes from Raj Sharma with B. Riley FBR.

--------------------------------------------------------------------------------

Rajiv Sharma, B. Riley FBR, Inc., Research Division - Analyst [30]

--------------------------------------------------------------------------------

Can you give us some color on the market-making this quarter as a proportion of total trading? How did it -- and how did it trend versus the volatility that you saw last quarter versus this quarter in the CVIX levels?

--------------------------------------------------------------------------------

Glenn H. Stevens, GAIN Capital Holdings, Inc. - President, CEO & Director [31]

--------------------------------------------------------------------------------

So just to confirm, you're talking about Q3 versus Q2?

--------------------------------------------------------------------------------

Rajiv Sharma, B. Riley FBR, Inc., Research Division - Analyst [32]

--------------------------------------------------------------------------------

Yes. Q3 versus Q2. Just how it -- how the amount of market-making you do versus unidirectional trade reflects -- I guess reflects some of the revenue capture. Can you talk about what proportion of the total trading was market-making? And if -- and also if your AI-related hedging model is helping you with the market-making proportions and the RPMs?

--------------------------------------------------------------------------------

Nigel Rose, GAIN Capital Holdings, Inc. - CFO [33]

--------------------------------------------------------------------------------

Raj, it's Nigel here. Are you saying how much of the activity we saw in Q3 was customers naturally offsetting each other versus us then trying to hedge the excess?

--------------------------------------------------------------------------------

Rajiv Sharma, B. Riley FBR, Inc., Research Division - Analyst [34]

--------------------------------------------------------------------------------

That's right.

--------------------------------------------------------------------------------

Nigel Rose, GAIN Capital Holdings, Inc. - CFO [35]

--------------------------------------------------------------------------------

We haven't disclosed that at this stage. I think in the Qs and Ks, we have that statement around. I think it's normally like 95%, 97% is either naturally hedged or hedged to brokers. And this quarter was no different to that.

--------------------------------------------------------------------------------

Glenn H. Stevens, GAIN Capital Holdings, Inc. - President, CEO & Director [36]

--------------------------------------------------------------------------------

And keep in mind that from a market-making perspective, if you will, you used the word unidirectional. I just want to clarify, there isn't any directional market-making within our business or within our capacity. So 100% of everything that we do is facilitating customer hedging. Now whether that's done, as we said, instantaneously or not -- just, look, just to clarify, the unidirectional aspect doesn't exist. It's -- but in terms of what portion, that remains pretty much high.

The difference here, most importantly, that drives our RPM capture, if you will, or our market-making piece is the propensity for customers to have 2-way trading. And what we mentioned, for example in Q1, when you had very narrow ranges but actually very well-defined ranges, you get very little bidirectional trading because it actually -- let's use your word in a different context, maybe.

You get a lot of unidirectional trading because when the euro is trading in a super tight range that repeats itself over a quarter, you get the preponderance of your customers all selling near the highs and all buying near the lows, which means our ability to capture offset -- our internalization is very low. So we end up having to do a lot of real-time external hedging, which means your spread capture is lower.

When you have less defined ranges and more 2-way trading, even with the same kind of volume, it's just a different nature, you end up with a 164 instead of a 50 in terms of an RPM. So I guess maybe if I phrase it that way, when you look back, it wasn't so much the nature of our market-making versus anything else, it was the nature of the customer trading that defines the difference between a 51 and a 164.

--------------------------------------------------------------------------------

Rajiv Sharma, B. Riley FBR, Inc., Research Division - Analyst [37]

--------------------------------------------------------------------------------

Okay. And also, the hedging model that you're talking about. Using AI, I see that the standard deviations of the daily revenues is going down. Does this help you in reducing the quarterly variability of RPMs going forward?

--------------------------------------------------------------------------------

Glenn H. Stevens, GAIN Capital Holdings, Inc. - President, CEO & Director [38]

--------------------------------------------------------------------------------

So it's good you asked the question because it's a little bit with a question that came up earlier in the screen today. The key part is what you said about quarterly because one of your colleagues said, well, hey, I see that the Sharpe ratio is improving because it's going higher over time and I see that the standard deviation of daily P&L is going down, that means improving because it's less variable. But why the hell did you have this big difference between 164 and 50?

And the first thing to highlight is to say, well, yes, in those 2 quarters, that -- those are variable. Number one, I'd point out that they go right smack to our long-term average of 105, when you add 164 and 50 and divide it by 2, well, pretty close. And then over a longer term, this AI model is not necessarily designed to reduce quarterly variability, it's designed to reduce overall variability, meaning over a year, over 16 months, over 2 years.

So we've been running these parallel testings since Q1 of '18. And by every measure, in the last, let's call it, 18 months, if you go from Q1 of '18 to now, so roughly 18 months, all of the measures show that we would have had a higher standard deviation of daily P&L and we would have had a lower Sharpe ratio if we [started] using our new model. So it has improved. And I guess the one thing that we can't show you is to say, oh, by the way, the difference between Q2 and -- no, sorry, Q3 of '18 and Q1 of '19 would have been even greater if we weren't running the model.

--------------------------------------------------------------------------------

Rajiv Sharma, B. Riley FBR, Inc., Research Division - Analyst [39]

--------------------------------------------------------------------------------

Got it. Got it. I understood. I'll take it offline now.

--------------------------------------------------------------------------------

Operator [40]

--------------------------------------------------------------------------------

And that does conclude the question-and-answer session as well as the call itself. The conference has now concluded. Thank you for attending today's presentation. You may now disconnect your lines.