Roundtable Discussion; The Future of Mineral Sands. Watch the video here.
Ghostrider re your comments on CGF and Mr Bates, I think there's something to be learned. Yes, he didn't know that the platform did not offer Gamestop, nor had it done for a year. Sell-side analysts are busy, loaded with work, and seldom have time to do any primary research. This sometimes leads to opportunity for people who do. With these businesses, I'd suggest that a tiny minority of institutional holders actually have used it and peers' trading apps.
continues...
As for current expectations, the company did $382m of customer income in 2019, with only 77k customers carried from 2018 into Q1 2019. 2019 had really low volatility.
This year it will start with c. 165k customers carried from 2020. It is expected to do $430m revenues after what I estimate to be a $22m hit from lower leverage in Australia, so $452m before that. Is it realistic that you spend $221m on marketing to drive so little growth over the low-vol 2019 level? Exceptionally improbable.
I have 2021 customer income of $690m, revenue of $590m. 55k new customers Q1, then a low 24k assumed in subsequent quarters. A low $980 quarterly ARPU is assumed, then lowered by ASIC measures. At 1408p the stock is on 6.3x my this year forecast of 223p. Consensus is at 144p.
It seems conservative to expect the company to trade to 10x, with these returns. If the stockbroking product and other non-trading product are well-received, and the moderate external hedging is understood to be reducing quarterly revenue volatility, then the multiple will be higher. Of those 10 million downloaders of the app from the Google store, perhaps some, who do not want to trade with leverage right now, might be customers in another way?
The CG price target is unchanged. It came from a 29 September note in which Mr Bates reckoned the business would do $526m revenues and 185p EPS in 2021 and $406mm revenues and 115p in 2022. He put a 5% discount on a long-run 8.7x multiple to value Plus500 at 8.3x that 115p.
By valuing the company on FY22 "normal" EPS he was ignoring the profits earned before that time. He had $148m PAT in FY22. He had Plus500 earning $272m PAT in FY21 and $529m in FY20. So that's $124m of "supernormal" profits in FY21 and $381m in FY20. They add up to $505m.
At a 958p price target with 105m shares and then a 1.286 FX rate at the time, that's a $1295m valuation of the company. The $505m of "supernormal" profit which he implicitly discards is 39% of his target price. That seems not great.
High churn was part of his reasoning. He estimated 45% churn in 2020, then 60% in 2021. The 2020 churn rate ended up being 30%. Somehow the CGF target price is unchanged. Curious.
He estimated 100k new clients in 2021. Let's see how Q1 pans out.
In the same report he values CMC at a 5% premium to IGG's 14.3x historical average 1y forward multiple. He uses CY22, but has CMC's March '22 and March '23 profits at the same level, because he believes it can recycle profits to create growth. This reminds me of another company.
The essence is that CGF has CMC's FY to March '23 profits at 45% of the level of FY21, while it has Plus500's 2022 profits at 34% of 2020. As ever, this is driven by an annual customer churn assumption. What he misses, despite having been the company's broker for years, is that all customers are not equal. The customers in any cohort who stop trading first are the ones with the shallower pockets. Yes, Plus500's customer churn is high compared to peers, but it has a huge number of lowish value customers from whom it makes good returns, having as it does a low cost online process and support function, and then it has 9% of customers who are > 90% of the revenues. As a result you would expect high churn in terms of customer numbers, but this has little bearing on the decay rate of a cohort's revenues.
Nowhere does he understand that the c. 300k new customers in 2020, the result in great part of $221m of marketing spend, will have lasting value. The company gave us cohort revenue data in August and again just now. It looks like the 2020 new customers only generated, at a maximum, $170m revenue in the year. They will generate that sort of number again this year, i.e. a lower run-rate but in 2020 they were on average only there for a little more than half of the year. Look at the 'High return on Marketing Investment" slide and you will see that that sort of year two revenue return on marketing spend is typical.
well found, Ghostrider
Low_PE_seeker - Just to try to help - it's not that simple. There's a massive skew in customer values, and the ones who stop trading first are the low value ones. E.g., if you recruit 10k new customers and 5k stop trading by a couple of quarters later, they were only 10-20% of the lifetime value of the cohort. As last disclosed, 9% of customers were 91% of revenues. At IG it's 10% who are 80%.
The better angle is to think of dollars rather than customers. When the company invests $1, it typically gets that back in the first year. At a 55-65% operating margin with only a 12% tax rate, that's already a great bottom-line return. They then get back another dollar in revenues in year 2 (+/-). There's a decline after that but typically in a 5Y period the company sees a 4-5x revenue return on marketing investment.
The above is why this year's huge marketing spend matters so much to 2021. To the best of my knowledge, no stockbroking analyst considers meaningfully the concept of return on marketing investment when forming their estimates.
14cr - It's my interpretation. If I say "the weather was lovely last week, and, in particular, Thursday", I'm saying that the characteristic I ascribe is greater or stronger for the subset. The statement said to me that CI was stronger than revenue.
I totally agree that yesterday is likely to have been a great day.
Cheers ggplyr. The statement says that revenue is also strong. The issue will be that the book is huge now and CTP can move around a fair bit. As a shareholder you should want this, in that the volatility of revenues is the small price you pay for the massive benefit of losing not a cent in external hedging costs. IG loses approximately a quarter of its gross revenues from customers to such costs.
The lovely thing about short-term sell side numbers is that 2020 numbers will have to go up even if CTP were, somehow, a very material number.
As for forecasting, I'd agree at least that the company could give a range of Customer Income forecasts. I'd note that the H1 presentation gives, for the first time, lifetime value evolutions of previous annual customer cohorts. That gives you a real steer on what sort of revenues a customer brings in Y1 and Y2 and beyond; the point being that it's remarkably knowable.
That makes sense. Don't forget however that the AGM statement draws a distinction between CI and revenue, so saying that CTP (market P&L) is negative, quarter-to-date. Though we may all be happy that this should tend to zero over time, and it's Customer Income that we care about, it was at that point nonetheless some sort of a drag on revenue.
I have $254m CI this quarter and $200m Q4. I have 45% customer churn this quarter, which is at the high end of what I'd expect, when you look at then customers arrived and what ARPUs have been running at since then. I have -$52m for customer trading performance, which I can barely forecast, but at least know was -ve QTD at the AGM.
With $200m revenue, $47m of advertising cost, assuming same AUAC as in Q2, and $46m of other costs, that would be $108m of EBITDA, which compares to $117m Liberum H2 expectation.
It seems pretty likely to me that the company hits FY EBITDA at the end of this quarter. The precise date will depend more on the ebb and flow of CTP.
Almost more important than any of the above is the continued high level of financial newsflow, sparking interest across the globe, and deposits for the World's Trading Machine. New customers who arrive this quarter will be generating high revenues in 2021. A lazy reader/ skier of Liberum or Peel Hunt Research might think this stock is on 10x 2021. It isn't.
yes, and to anchor it back to the expectations of $193m revenue in H2 set by Liberum. Peel Hunt at $240m.
ggplr - I don't think there's any ambiguity. The active customer number is any customer who traded in the period, hence the number for H1 2020 is not the sum of Q1 and Q2, because some customers traded in both quarters, but count once only in the H1 number. The same principle applies to annual actives.
You are right to consider the language closely. The company says "The operational momentum achieved during H1 2020 has continued into the second half to date, with Plus500 making excellent progress across all key commercial and financial performance metrics."
In my opinion, you could not say this if there had been a deterioration in any single one of the operational metrics since H1. These metrics include new customers, active customers, AUAC, ARPU, Customer Income, trades and deposits, as given below on slide 9 of the H1 results presentation, where it lists metrics, saying:
"Outstanding performance across all metrics:
Heightened volatility in unprecedented market conditions
High quality and consistent performance of Plus500’s leading-edge technology platform
Unprecedented levels of New Customers1 and Active Customers2at attractive ARPU3and AUAC4
AUAC was materially lower at $634 (H1 2019: $1,079)
Record levels of Customer Income5of $556.9 million, up 218%
Over 47 million customer trades (H1 2019: 17.5 million)
Client deposits of $1,653.4million (H1 2019: $467.1 million)"
Active customers are up sequentially from Q2. If churn was at the level of Q2, 23%, then 61k new customers would have to have deposited money. For new customers to be running ahead of this rate is remarkable. Remember that the company brought 91k new customers in all of 2019, but 198k in H1. Customers trading for the first time in Q3 will be generating high revenues into 2021. Look at slides 19 and 21 of the H1 presentation "Long term customer cohort value", and do the maths on the revenues that are generated in 2021 from the c. $200m of marketing spend that will be invested in 2020. The investment will be more than double the $96m of 2019. It was $126m just in H1.
There was a note from Peel Hunt yesterday (corporate relationship with CMC) which suggested that Plus500 new customers in 2021 would be 61k. Plus500 was attracting numbers above that level even back in 2014. That is the level of institutional understanding of this stock, and the great opportunity here.
14cr and Mark - Market moves here:
http://www.filedropper.com/marketmoves20200730>http://www.filedropper.com/download_button.png width=127 height=145 border=0/>
Robert - hi- I wasn't saying that the share price wouldn't go down on the day ASIC announces lower leverage, but I am saying that estimates will change minimally, and that as a result the share price a week later will be unaffected.
This is, after all, a consultation that closed late last year, the leverage levels have been announced, then the regulator said consultations were on hold for now.
ASIC is in the numbers. The company guided to a 65% impact on 15% of their business, and it went into Liberum forecasts. This has recently been reasserted in e.g. the note of 16 March: "Given the exceptional nature of the environment during Q1’20, we expect a significant step down in subsequent quarters, as shown in Fig 5. This dynamic also reflects the impact of the regulatory change announced in Australia last year. On an underlying basis we estimate that the full year impact of this
change is c$35m – consistent with the loss of 65% of the revenues in country."
I follow your thinking and agree with what you say about the first two months of the year.
It's amazing to a pull up a 5 year chart of the VIX and see for how long that low level of volatility was considered to be normal. When an analyst forecasts 2021 at $400m, or even lower, not only have they totally forgotten the whole concept of return on marketing investment, but they are also implicitly assuming 2019 levels of volatility which, for all the reasons you cite, might very well not happen.
hello. I'm not sure you really believe that the current customer base will produce customer income at this level all the way through 2020 and into 2021, but isn't that point that we do not have to believe anything so positive? The betting line is a set of H2 revenue estimates of between $195 and $250m. Given that customer income will be something like $315-340m in Q2, that is staggering. The probability that the business slows down to that extent is infinitesimal.
Over time > 98% of Plus500 revenues will come from spreads and overnight charges. Because they don't hedge, no customer revenues go to hedging costs, but revenues are more volatile than for example IG. IG reported net trading revenue of £477m in FY19, but this was after £61m of hedging costs and £245m of hedging losses. Client P&L was £139m to the benefit of the client. Add them up and client income was £644m, i.e. IG lost 26% of client income to hedging costs.
From the start, Plus500 has been set up not to make this mistake. Shareholders in Plus500 get all the cashflows, with no leakage, but they do have to accept higher short-term revenue volatility. The staggering thing is that this business has seen negative $147m of customer trading performance so far in Q2, and yet is still on track to hit Liberum expectations of $732m revenues for the year.
Robert, hello, if I may, that's not the case, the company is 100% clear they they do not hedge, and has been clear since IPO. It's all described again in the prospectus for the move to the main market. What they do not disclose is the split of the book by asset class or instrument, saying that it changes all the time, anyway. You can look at the platform and see that customers are almost always very long equities, such that Plus500 is short the market. This has always been the case, and most of the time spreads are generated at such a rate that customer P&L does not matter, but when you get a big move, this is no longer the case. That's what you saw in a small way in Q1'19, and in a big way just now.
For revenues, Liberum is at $200m, Berenberg $294m and Autonomous must be at close to $300m as he's $864m for the full year. I expect $340m of customer income. Customer P&L is unforecastable, though we know it was a substantial number by late April.
Such revenues mean that a 60% interim dividend would be c. 180p a share, which is rather significant.
More important, however, is the marketing investment, which will drive FY20 and FY21 revenues. It was $53m in Q1. Liberum went to $50m for Q2, which means it will be a record quarter, because, if the company knew it would spend at least $50m in Q2 by 28 April, it weill have done at least that numberThere were 83k new customers in Q1, and I'd expect this quarter to be bigger. If the marketing machine is applying $65m of spend this quarter, which is possible, then the company grows in a sustainable way. You will see customer churn in future quarters, but you have to remember that the customers who trade with the company for 6 months are not the valuable customers. It's the ones with deeper pockets who love to trade and will continue to do so who matter, hence, while you might see customer churn rise in Q3, it does not mean that the ROI on the elevated marketing spend is not at least as impressive as it always has been.
Don't forget that IG had a multi-hour system failure this quarter, and an enduring inability to handle the new customer applications, and to process withdrawals in good time. All this will have contributed to Plus500's further market share gains.
IG: £665
OliG - the company has not made a forecast. It does however guide the one of its two joint brokers who writes research on the company. The company has previously, at the time of results, referred to the Liberum number as being the result of their guidance.