BUR.L - Burford Capital

[Guest Schwab711]

I meant VRX in the sense that no matter how times it was written that inconsistent numbers, misleading narratives, ect are not a good sign, there was always an excuse for the inconsistencies to explain why it was fine. I didn't mean to say that BUR is VRX (or that it is a fraud or anything like that). The investment thesis seems as unbreakable as VRX's was from my point-of-view (which was correct in the case of UBNT and I was wrong on the importance of the issues I saw).

 

I guess to change gears, how are folks valuing BUR? What's the thesis that doesn't rely on P/E multiple? Has anyone taken a stab at a DCF?

[Normax59]

I meant VRX in the sense that no matter how times it was written that inconsistent numbers, misleading narratives, ect are not a good sign, there was always an excuse for the inconsistencies to explain why it was fine. I didn't mean to say that BUR is VRX (or that it is a fraud or anything like that). The investment thesis seems as unbreakable as VRX's was from my point-of-view (which was correct in the case of UBNT and I was wrong on the importance of the issues I saw).

 

I guess to change gears, how are folks valuing BUR? What's the thesis that doesn't rely on P/E multiple? Has anyone taken a stab at a DCF?

 

Not really sure how we would find FCF given their cash flow statement. IMF has their own FCF calculation in their annual report that could be applied though.

 

Net Profit After Tax

+ Litigation Contract Expenses (would need to determine the equivalent for BUR)

+ Depreciation

+ Change in WC

+ Net Investments (New investments - receipts)

- Capex

= Free cash flow

[Jerry Capital]

You can't think of it like that.

 

You have to be more creative.

 

[Guest Schwab711]

You can't think of it like that.

 

You have to be more creative.

 

What's the point of being cryptic? You are long. What formula/process did you use?

[Jerry Capital]

It's not cryptic.

 

I don't have a formula or calculation.

 

It's more about looking out a long time horizon, and thinking, can they take what might have been a lucky win with Peterson and build some sort of moat, shift to a more recurring revenue model with an asset management arm and portfolio financing.

 

The multiple that a truly uncorrelated (not negative but zero correlation) equity with stable earnings would be incredibly high (think $V $MA type mutliple)...

 

You have decent downside protection IMO and a very interesting bull case.

 

It's more of a Bill Miller style investment for me, and that is reflected in the weighting in my portfolio.

 

Does that make sense?

[Normax59]

Here's as creative as I can get:

 

2018

 

Net Profit before Tax: 305,114

+Adjustments For Non-Cash: 16,709

+Proceeds from Investments: 629,410

-Funding of Investments: 738,243

+Proceeds from new initiative investments: 8,757

- Funding of new initiative investments: 33,074

+ Proceeds from Financial Liabilities: 74,044

-Due from brokers and due to brokers: 75,566

- Change in Operating WC: 3,598

- Purchases of Fixed Assets: 650

 

== 182,636

 

2017

 

Net Profit before Tax: 249,181

+Adjustments For Non-Cash: 13,113

+Proceeds from Investments: 363,889

-Funding of Investments: 569,564

+Proceeds from new initiative investments: 2,623

- Funding of new initiative investments: 6,467

+ Proceeds from Financial Liabilities: 36,510

-Due from brokers and due to brokers: 41,678

+ Change in Operating WC: 4,988

- Purchases of Fixed Assets: 104

 

== 52,491

 

2016

 

Net Profit before Tax: 104,053

-Adjustments For Non-Cash: 46,971

+Proceeds from Investments: 180,772

-Funding of Investments: 275,698

+Proceeds from new initiative investments: 13,135

- Funding of new initiative investments: 4,274

+ Proceeds from Financial Liabilities: -

-Due from brokers and due to brokers: -

+ Change in Operating WC: 5,868

- Purchases of Fixed Assets: 1,570

 

== -24,557

 

2015

 

Net Profit before Tax: 67,877

-Adjustments For Non-Cash: 2,899

+Proceeds from Investments: 140,196

-Funding of Investments: 91,932

+Proceeds from new initiative investments: 6,196

- Funding of new initiative investments: 21,265

+ Proceeds from Financial Liabilities: -

-Due from brokers and due to brokers: -

+ Change in Operating WC: 6,829

- Purchases of Fixed Assets: 421

 

== 104,581

2014

 

Net Profit before Tax: 47,324

+Adjustments For Non-Cash: 4,615

+Proceeds from Investments: 63,010

-Funding of Investments: 91,022

+Proceeds from new initiative investments: 2,504

- Funding of new initiative investments: 2,821

+ Proceeds from Financial Liabilities: -

-Due from brokers and due to brokers: -

+ Change in Operating WC: 790

- Purchases of Fixed Assets: 100

 

== 24,300

 

I'm willing to hear suggestions for changes.

[Liberty]

New post (part 1 of 2) on Burford by Scuttleblurb (subscription required):

 

https://www.scuttleblurb.com/bur1/

[peterHK]

You can't think of it like that.

 

You have to be more creative.

 

What's the point of being cryptic? You are long. What formula/process did you use?

 

I don't actually think this is that hard. The key variables are what level returns stabilize at, and when. For that, you can run a sensitivity analysis on stabilized ROE vs. COE in however many years to get a P/BV multiple, and then work backwards: ROE x BV = $X of BV accretion per year etc.

 

E.g if ROE is 30% for 3 years and they reinvest all their earnings (yes a simplified assumption, but historically they have been able to reinvest OVER 100%), then 3 year out BV is $13.68/share.

 

If after that ROE drops to 15% vs. a COE of 10% and they never grow (again, an unreasonable assumption, but a conservative one), your justified P/B is 1.5. 1.5x 13.68 = $20.53/share.

 

Repeat the above with ROE at today's levels and dropping at various rates for 2,3,4,5 etc. years and you can get a pretty good idea of the range of outcomes.

 

Then ask yourself: how quickly will returns fall? Why might they NOT fall? How fast do they have to fall in order for me to lose money on this investment? Does it make sense that a business goes from earning 30% ROE's to 20% in 1 year? In 2 years? What competitive forces are going to make that happen in that short of a timeframe?

 

You cannot get the degree of accuracy with Burford that you can with many others; ROE can be lumpy, cash ROE is certainly lumpy depending on realizations. That doesn't mean you can't construct a range of outcomes, figure out which set (bull/bear/middle) look qualitatively more logical, and then construct the risk/reward on the investment.

 

My own view is that I find it very hard to believe that returns fall dramatically in the next 2-3 years, and because of that, the risk of a permanent loss of capital outside accounting fraud I feel is very low.

 

 

 

 

[Guest Schwab711]

Returns might not immediately fall if Petersen is successful. The implied odds of Petersen success is estimable based on secondary transactions made by sophisticated investors and Jeffries' estimate of the Petersen payoff, assuming success.

 

It is almost certain that one of the following is true: (1) BUR made a really bad decision selling their interest in Petersen (the true odds of success aren't so low); (2) the true expected outcome of Petersen, assuming success is much lower than $2.5b estimated by Jeffries; or (3) BUR is much riskier than you think.

 

Outcome #2 is probably closest to your "30% for a few years and falling". In that case, BUR will do fairly well. There are many outcomes where BUR will fall precipitously. I think the R/R skew for BUR is the reverse of what many assume. More Petersen-like cases is the only way to support 3x+ BV.

[peterHK]

Do ppl bitch this much about unrealized gains when their PE firm marks up the value of their investment "becuz DCF"?

 

Take a look at some PE financials, let me know if you find one that has mark to market revenue represented 65% by unrealized gains. Unrealized gains are so minute for them it’s not even an equivalent comparison. And if they did I would bitch about it.

 

The idea that the secondary market represents some sort of all knowing entity that is appropriate to mark your investments to is questionable. Yes, Burford's marks are significantly below market, unlike BPY, but I think the principle stands (at least for me): we know their marks are conservative... [snip]

 

 

 

Further, I'd look at IAS 37 (https://www.iasplus.com/en/standards/ias/ias37)  as a sort of inversion for Burford: you mark litigation liabilities based on estimates and based on specific events happening. One marks liabilities similar to how Burford marks assets, which makes intuitive sense because Burford is the other side of those liabilities. This isn't a real estate asset where we can get this year's NOI, cap it, and boom, you have an asset value. These are all contingent assets. My reading of Burford holding this below the most recent sale price is that they have a different view of the risk than the market does.

 

Further, on contingent assets, one doesn't even recognize them AT ALL, until the recognition of income is virtually certain. One could argue Burford should account for cost only, and realize all gains on receipt of cash. I'd be fine with that tbh.

 

 

I'd be willing to bet that there will be A LOT of questions on the call about it, and a lot of discussion in the annual report on it.

 

 

The first part is literally the point of FV accounting. It's to mark positions at what the market would pay (not you, me, the company, or otherwise - what the mark would take the asset for). That's why secondary transactions (especially ones where you [the company]) are personally party to matter so much.

 

Second, IAS 37 specifically says financial instruments are exempt from the accounting standard you bring up. Litigation claims can be seen as roughly equivalent to CDS contracts. They have negative carry (expenses of case or on-going premiums for CDS) and have a binary payoff. The CDS has a known payoff vs unknown. In that case we can look at European call options, which have an unknown binary payoff. Both Euro calls and CDS are marked to FV greater than $0 without complaint. These are often illiquid markets and still marked to secondary transactions. I think BUR is misleading people with the idea of overly conservative accounting. At some point, it allows BUR to smooth results that by their nature aren't smooth (I'm repeating myself at this point but this feels like VRX so I'm not surprised).

 

Third, I agree and hope that this is addressed on the call. I understand strategically why BUR doesn't want to state their marks but that's the downside of taking outside money.

 

 

 

In other examples of 'conservative' being thrown around in potentially misleading ways, BUR said in the 11/12/18 investor day presentation that:

Only two investments that were written up, amounting to 0.2% of total write-ups by dollar value, have ever turned into a loss

 

First, this comment implies the loss was roughly $1m or so. It's a weird stat. Anyway, in the 2018 AR, BUR noted that one claim that was previously written up was twice written down 2 years ago and in 2H 2018, which caused a "fair-sized" loss. So while that stat was correct on 11/12/2018 about completed investments, it's misleading. It looks like BUR writes down multiple claims every semi-annual period.

 

 

On 7/25/2018, BUR noted in their 1H 2018 investor presentation regarding Petersen:

We sold 3.75% of our entitlement for an effective

cash price of $30 million, implying a valuation of

$800 million for the original total Petersen

entitlement.

 

We carry our Petersen investment at a

lower carrying value than that for the

reasons we have enunciated previously.

 

Obviously Petersen wasn't marked at an implied valuation of $800m at 6/30/2018 if the sale supporting that valuation occurred on 7/11/2018. What BUR wrote isn't wrong. It can even be read as an appropriate comment. It's about the method of communicating though. BUR is clearly run by lawyers that know how to spin a story.

 

 

I'm not trying to pick on you, Peter. It just feels like the narrative of BUR doesn't match the financial picture so i'm trying to attack the narrative.

 

I don't think CDS is the best analogue to litigation, but Euro calls are a good one. Burford doesn't mark their cases to $0, they mark them to where they feel is fair value, much as one would an undervalued euro call. How do you value a Euro call on a private asset that seldom trades, for instance? How do you measure the volatility input? All of these are assumptions that are part of valuing level 3 assets, which is what Burford's cases are. Again, they've said that sales ARE inputs, but not the only one. I am fine with this, I regularly aknowledge that my public market equity's intrinsic value does not move around as much as the share price. I'm 99% sure that the value of litigation does not move around as much as what the current market prices for it do.

 

I would see the issue with smoothing if they did not have a track record of cash realization. The issue with VRX was that the cash flows did not track in any sense to their "cash EPS" measure, and cash ROIC was consistently very low. They also shifted segments around all the time to obfuscate organic growth. IMO Burford is a mirror image of a lot of this: we can see they have cash realizations and what those are; we can see that cash ROIC on investments is high; and we can see they go to immense lengths to explain and give us information in their filings.

 

On claims being written down, the full quote was the following:

 

"What investors will note, however, is the reduction by a couple

of percentage points in the final two years before

conclusion (now 33% and 10% versus 35% and 12%

as reported in the 2018 interim report), which is

entirely due to the impact of a fair-sized investment

that ultimately lost and which we wrote down

partially two years before its conclusion and again

during the year of conclusion, which when netted

against valuation increases caused the numbers to

decline somewhat – but also proves the point that

we change valuations in both directions"

 

I feel like owning up to this when the facts changed, and adjusting their numbers is not disingenuous or misleading, it's good disclosure. You also have to separate writing down on cost and writing down after writing up. Burford's claim is that it doesn't tend to write down cases it has first written up, as evidence of their conservative valuation. Again, the above case was a change to that fact, which they owned up to, disclosed and explained.  What Burford does not claim is that they don't write down cases held at cost. These result in the majority of their write downs.

 

I know you're not picking on me, but perhaps I'm biased in this, it seems like you're willingly picking at things rather than looking at what Burford does (e.g disclosing changes as per the above) right relative to peers in this industry. We can quibble on accounting all day long, but from the hundreds of reports I've read across industiries, I have seldom come across a management team that provides this much explanation and discussion in their reports. One view is that this is because they are good people trying to get us to understand the business; the other is that they are trying to explain away all the criticism. Which side you are on depends, I think, more on what type of person you are than the preponderance of evidence, because there is plenty of evidence supporting both sides.

[Normax59]

You can't think of it like that.

 

You have to be more creative.

 

What's the point of being cryptic? You are long. What formula/process did you use?

 

I don't actually think this is that hard. The key variables are what level returns stabilize at, and when. For that, you can run a sensitivity analysis on stabilized ROE vs. COE in however many years to get a P/BV multiple, and then work backwards: ROE x BV = $X of BV accretion per year etc.

 

E.g if ROE is 30% for 3 years and they reinvest all their earnings (yes a simplified assumption, but historically they have been able to reinvest OVER 100%), then 3 year out BV is $13.68/share.

 

If after that ROE drops to 15% vs. a COE of 10% and they never grow (again, an unreasonable assumption, but a conservative one), your justified P/B is 1.5. 1.5x 13.68 = $20.53/share.

 

Repeat the above with ROE at today's levels and dropping at various rates for 2,3,4,5 etc. years and you can get a pretty good idea of the range of outcomes.

 

Then ask yourself: how quickly will returns fall? Why might they NOT fall? How fast do they have to fall in order for me to lose money on this investment? Does it make sense that a business goes from earning 30% ROE's to 20% in 1 year? In 2 years? What competitive forces are going to make that happen in that short of a timeframe?

 

You cannot get the degree of accuracy with Burford that you can with many others; ROE can be lumpy, cash ROE is certainly lumpy depending on realizations. That doesn't mean you can't construct a range of outcomes, figure out which set (bull/bear/middle) look qualitatively more logical, and then construct the risk/reward on the investment.

 

My own view is that I find it very hard to believe that returns fall dramatically in the next 2-3 years, and because of that, the risk of a permanent loss of capital outside accounting fraud I feel is very low.

 

The calculation without unrealised gains in the income statement or balance sheet.

 

EDIT:

 

The first two pictures was what was original displayed.

 

I did an awful job when first posting this and didn't even give any context to what I was showing and quite frankly didn't even use the right numbers.

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

 

First - Unrealized gains for the last 4 years as reported on the balance sheet in tandem with equity ex. any unrealized gains along with the reported breakdown of income by realized and unrealized gains is shown below in the screenshot.

 

The average return ex unrealized gains over this period is 12.8%, without 2016 added in the mix this seems like a pretty stable number to project into the future. The large increase in total equity ex URG can be largely explained by the issuance of shares in the beginning of the year, removed the average increase y/y is 12.9%. BUR therefore needs to obtain a 14.68% return on previous years equity less URG's. BV less unrealized gains is shown over the next 3 years in screenshot 3 under the assumption BUR can continue to grow it at the rates they have over the last 4 years.

 

We can now make some conclusions on what the income statement and BV would need to be composed of at the end of 2019 if they were to report a 30% ROE on previous years equity based on your calculation. Net Income would come in at around 400-408MM composed of 108MM in realized gains and 300MM in unrealized gains. Given the recent Petersen sale in July this assumption doesn't seem outlandish and in fact might a very conservative estimate

 

The balance sheet would have 1.7BB in equity with 934MM in unrealized gains. Investments as a % of equity has settled around 110-130% of equity so we can estimate total investments between 1.87-2.21BB Y19. Similar conclusion's can be drawn from the rest of the screenshot.

 

All this is under the assumption that they will refrain from issuing equity, something that I feel they will need to continue to do in the future as it seems their high growth has been completely dependent on outside capital far more than investment proceeds. 

 

End of year 3 we get your fair value of $20.55 its also implied that BUR would have 1.9BB in unrealized gains that the market has valued at $12.87 a share or 63% of the fair value.

 

I'm leaving the spreadsheet as well at the bottom, if anyone wants to change the assumptions or give suggestions or that you hate it.

BUR_P_to_BV_Test.xlsx

[peterHK]

Returns might not immediately fall if Petersen is successful. The implied odds of Petersen success is estimable based on secondary transactions made by sophisticated investors and Jeffries' estimate of the Petersen payoff, assuming success.

 

It is almost certain that one of the following is true: (1) BUR made a really bad decision selling their interest in Petersen (the true odds of success aren't so low); (2) the true expected outcome of Petersen, assuming success is much lower than $2.5b estimated by Jeffries; or (3) BUR is much riskier than you think.

 

Outcome #2 is probably closest to your "30% for a few years and falling". In that case, BUR will do fairly well. There are many outcomes where BUR will fall precipitously. I think the R/R skew for BUR is the reverse of what many assume. More Petersen-like cases is the only way to support 3x+ BV.

 

I agree, the reason Burford likely sold is that Petersen expected value is less than $2.5bn. Nobody here is underwriting Burford using $2.5bn as the value for Petersen.

 

Nobody is arguing this will trade at 3x BV in 5 years. I underwrite it at 1.5x in 5 years just FYI. I think you haven't done the work if you think that Petersen repeats is the only way this business can work (hint hint $100+mn portfolio investments that generate mid 20 IRR's). You do not need 30% ROE's to continue indefinitely for this investment to work; in fact, you don't need 30% ROE's to continue for more than ~3 years for this investment to work if you look at the other lines of business Burford is in and what their TAM is.

C

[peterHK]

You can't think of it like that.

 

You have to be more creative.

 

What's the point of being cryptic? You are long. What formula/process did you use?

 

I don't actually think this is that hard. The key variables are what level returns stabilize at, and when. For that, you can run a sensitivity analysis on stabilized ROE vs. COE in however many years to get a P/BV multiple, and then work backwards: ROE x BV = $X of BV accretion per year etc.

 

E.g if ROE is 30% for 3 years and they reinvest all their earnings (yes a simplified assumption, but historically they have been able to reinvest OVER 100%), then 3 year out BV is $13.68/share.

 

If after that ROE drops to 15% vs. a COE of 10% and they never grow (again, an unreasonable assumption, but a conservative one), your justified P/B is 1.5. 1.5x 13.68 = $20.53/share.

 

Repeat the above with ROE at today's levels and dropping at various rates for 2,3,4,5 etc. years and you can get a pretty good idea of the range of outcomes.

 

Then ask yourself: how quickly will returns fall? Why might they NOT fall? How fast do they have to fall in order for me to lose money on this investment? Does it make sense that a business goes from earning 30% ROE's to 20% in 1 year? In 2 years? What competitive forces are going to make that happen in that short of a timeframe?

 

You cannot get the degree of accuracy with Burford that you can with many others; ROE can be lumpy, cash ROE is certainly lumpy depending on realizations. That doesn't mean you can't construct a range of outcomes, figure out which set (bull/bear/middle) look qualitatively more logical, and then construct the risk/reward on the investment.

 

My own view is that I find it very hard to believe that returns fall dramatically in the next 2-3 years, and because of that, the risk of a permanent loss of capital outside accounting fraud I feel is very low.

 

The calculation without unrealised gains in the income statement or balance sheet.

 

So you don't see an issue with comparing realized gains from some cases to the equity in ALL cases, both concluded and unconcluded. 

 

If you don't see how this is wrong, then I can't help you.

[Guest Schwab711]

Returns might not immediately fall if Petersen is successful. The implied odds of Petersen success is estimable based on secondary transactions made by sophisticated investors and Jeffries' estimate of the Petersen payoff, assuming success.

 

It is almost certain that one of the following is true: (1) BUR made a really bad decision selling their interest in Petersen (the true odds of success aren't so low); (2) the true expected outcome of Petersen, assuming success is much lower than $2.5b estimated by Jeffries; or (3) BUR is much riskier than you think.

 

Outcome #2 is probably closest to your "30% for a few years and falling". In that case, BUR will do fairly well. There are many outcomes where BUR will fall precipitously. I think the R/R skew for BUR is the reverse of what many assume. More Petersen-like cases is the only way to support 3x+ BV.

 

I agree, the reason Burford likely sold is that Petersen expected value is less than $2.5bn. Nobody here is underwriting Burford using $2.5bn as the value for Petersen.

 

Nobody is arguing this will trade at 3x BV in 5 years. I underwrite it at 1.5x in 5 years just FYI. I think you haven't done the work if you think that Petersen repeats is the only way this business can work (hint hint $100+mn portfolio investments that generate mid 20 IRR's). You do not need 30% ROE's to continue indefinitely for this investment to work; in fact, you don't need 30% ROE's to continue for more than ~3 years for this investment to work if you look at the other lines of business Burford is in and what their TAM is.

C

 

You are misinterpreting the meaning of the tables. $2.5b isn't the expected value, it's the estimated proceeds if Petersen is successful. Petersen sold presumably because they think the odds of success are less than [see table], assuming X years to completion and Y% return hurdle.

 

 

 

My point in all of this is that all of the issues I'm raising are interconnected.

 

The accounting matters because we need to know if Petersen is reflected in BV/earnings or not and to what degree. Then we look at the implied odds of success for Petersen and we can see that there may be material downside risk to BUR's future earnings. Future above-normal returns are expected to be supported by the remaining Petersen recognition and cases related to the Petersen outcome not previously marked up, to our knowledge, but are expected to be material. If Petersen is not successful, BV and future earnings will decline. The accounting determines if this is a ST hit (BV decline) or LT hit (earnings not meeting expectations for many periods). Now that we agree that write-downs are common for cases held at cost, you can see why I think Petersen represents most of BUR's historical unrealized earnings.

 

If Petersen is not successful, it's possible BUR will no longer be the largest litigation finance entity since a new entity raised $2b in the last 2 years. That would change the thesis. Their lower returns may affect their ability to raise capital for future funds. The odds of such a monumental shift in financials and sentiment is not reflected in any of the investment pitches that I've seen.

 

There's potentially a 50% chance that BUR faces a 30%+ decline in BV and has materially lower prospects of high returns (basically similar to IMF after their 2008-2012 high-return period, which was buoyed by a particularly good investment). That's why I'm raising all these points. Is it a good bet to coin flip material permanent capital loss for outperformance? Maybe, I don't know. You may very well be correct on BUR and I really do hope you are (I can't and probably won't short BUR so what do I care). As I said, I like the industry.

 

 

I don't really know what the bull thesis looks like. No one has actually put numbers to the narrative. Jerry says it's 30x+ some number. You make a seemingly reasonable argument that we have clarity to the current price and then you like the industry as a whole. I disagree on the odds that BV reaches today's price in a few years but it's always easier to nitpick someone else's valuation. Said another way, every valuation has weaknesses.

 

I've once tried to write how I think BUR should be valued but I'm not sure how well I actually did explaining my thoughts. I would separate out extraordinary cases like Petersen and look at the remaining income stream. Depending on the returns on the more repeatable stream, I would probably arrive at a value of 1x-2x BV. Presumably the asset management income and the like increases the returns and increases BV. I would value Petersen or similar at their expected value (which may or may not be a multiple of their BV).

[Normax59]

You can't think of it like that.

 

You have to be more creative.

 

What's the point of being cryptic? You are long. What formula/process did you use?

 

I don't actually think this is that hard. The key variables are what level returns stabilize at, and when. For that, you can run a sensitivity analysis on stabilized ROE vs. COE in however many years to get a P/BV multiple, and then work backwards: ROE x BV = $X of BV accretion per year etc.

 

E.g if ROE is 30% for 3 years and they reinvest all their earnings (yes a simplified assumption, but historically they have been able to reinvest OVER 100%), then 3 year out BV is $13.68/share.

 

If after that ROE drops to 15% vs. a COE of 10% and they never grow (again, an unreasonable assumption, but a conservative one), your justified P/B is 1.5. 1.5x 13.68 = $20.53/share.

 

Repeat the above with ROE at today's levels and dropping at various rates for 2,3,4,5 etc. years and you can get a pretty good idea of the range of outcomes.

 

Then ask yourself: how quickly will returns fall? Why might they NOT fall? How fast do they have to fall in order for me to lose money on this investment? Does it make sense that a business goes from earning 30% ROE's to 20% in 1 year? In 2 years? What competitive forces are going to make that happen in that short of a timeframe?

 

You cannot get the degree of accuracy with Burford that you can with many others; ROE can be lumpy, cash ROE is certainly lumpy depending on realizations. That doesn't mean you can't construct a range of outcomes, figure out which set (bull/bear/middle) look qualitatively more logical, and then construct the risk/reward on the investment.

 

My own view is that I find it very hard to believe that returns fall dramatically in the next 2-3 years, and because of that, the risk of a permanent loss of capital outside accounting fraud I feel is very low.

 

The calculation without unrealised gains in the income statement or balance sheet.

 

So you don't see an issue with comparing realized gains from some cases to the equity in ALL cases, both concluded and unconcluded. 

 

If you don't see how this is wrong, then I can't help you.

 

Their investment weighted average duration is 1.8 years and you wanted me to exclude the ongoing investment equity that is going to produce the majority of realized NI gains in year 2 and 3?

[peterHK]

Returns might not immediately fall if Petersen is successful. The implied odds of Petersen success is estimable based on secondary transactions made by sophisticated investors and Jeffries' estimate of the Petersen payoff, assuming success.

 

It is almost certain that one of the following is true: (1) BUR made a really bad decision selling their interest in Petersen (the true odds of success aren't so low); (2) the true expected outcome of Petersen, assuming success is much lower than $2.5b estimated by Jeffries; or (3) BUR is much riskier than you think.

 

Outcome #2 is probably closest to your "30% for a few years and falling". In that case, BUR will do fairly well. There are many outcomes where BUR will fall precipitously. I think the R/R skew for BUR is the reverse of what many assume. More Petersen-like cases is the only way to support 3x+ BV.

 

I agree, the reason Burford likely sold is that Petersen expected value is less than $2.5bn. Nobody here is underwriting Burford using $2.5bn as the value for Petersen.

 

Nobody is arguing this will trade at 3x BV in 5 years. I underwrite it at 1.5x in 5 years just FYI. I think you haven't done the work if you think that Petersen repeats is the only way this business can work (hint hint $100+mn portfolio investments that generate mid 20 IRR's). You do not need 30% ROE's to continue indefinitely for this investment to work; in fact, you don't need 30% ROE's to continue for more than ~3 years for this investment to work if you look at the other lines of business Burford is in and what their TAM is.

C

 

You are misinterpreting the meaning of the tables. $2.5b isn't the expected value, it's the estimated proceeds if Petersen is successful. Petersen sold presumably because they think the odds of success are less than [see table], assuming X years to completion and Y% return hurdle.

 

 

 

My point in all of this is that all of the issues I'm raising are interconnected.

 

The accounting matters because we need to know if Petersen is reflected in BV/earnings or not and to what degree. Then we look at the implied odds of success for Petersen and we can see that there may be material downside risk to BUR's future earnings. Future above-normal returns are expected to be supported by the remaining Petersen recognition and cases related to the Petersen outcome not previously marked up, to our knowledge, but are expected to be material. If Petersen is not successful, BV and future earnings will decline. The accounting determines if this is a ST hit (BV decline) or LT hit (earnings not meeting expectations for many periods). Now that we agree that write-downs are common for cases held at cost, you can see why I think Petersen represents most of BUR's historical unrealized earnings.

 

If Petersen is not successful, it's possible BUR will no longer be the largest litigation finance entity since a new entity raised $2b in the last 2 years. That would change the thesis. Their lower returns may affect their ability to raise capital for future funds. The odds of such a monumental shift in financials and sentiment is not reflected in any of the investment pitches that I've seen.

 

There's potentially a 50% chance that BUR faces a 30%+ decline in BV and has materially lower prospects of high returns (basically similar to IMF after their 2008-2012 high-return period, which was buoyed by a particularly good investment). That's why I'm raising all these points. Is it a good bet to coin flip material permanent capital loss for outperformance? Maybe, I don't know. You may very well be correct on BUR and I really do hope you are (I can't and probably won't short BUR so what do I care). As I said, I like the industry.

 

 

I don't really know what the bull thesis looks like. No one has actually put numbers to the narrative. Jerry says it's 30x+ some number. You make a seemingly reasonable argument that we have clarity to the current price and then you like the industry as a whole. I disagree on the odds that BV reaches today's price in a few years but it's always easier to nitpick someone else's valuation. Said another way, every valuation has weaknesses.

 

I've once tried to write how I think BUR should be valued but I'm not sure how well I actually did explaining my thoughts. I would separate out extraordinary cases like Petersen and look at the remaining income stream. Depending on the returns on the more repeatable stream, I would probably arrive at a value of 1x-2x BV. Presumably the asset management income and the like increases the returns and increases BV. I would value Petersen or similar at their expected value (which may or may not be a multiple of their BV).

 

1) Where you pulling 50% from?

 

2) Expected value: https://en.wikipedia.org/wiki/Expected_value

 

I give up.

[Guest Schwab711]

50% is roughly the middle of the table I posted.

 

I would correct you on E(X) but it's obvious and I'm starting to like that you have no idea what you own.

[no_free_lunch]

Well written, thorough investment thesis on Burford.

 

https://millonariosanonimos.com/burford-capital-ltd-investment-thesis/

 

The section on prestige and relationship as a moat is worth considering.  I think it was mentioned earlier in this thread as well.  Maybe a prospective corporation doesn't want to reveal it's litigation details to the whole world, so they will limit disclosure to law firms they trust.  However, there are also a lot lawyers out there so I am not sure how deep this moat is.

 

To get investments you must have a very good relationship with law firms and potential clients. To study an investment, you have to make a complete due diligence, and the customer has to grant you access to very critical and confidential information. As a result, the industry will never become an open auction in which everybody can get access to that information. That makes the relationship with Burford very sticky. Nobody has been fired for working with Burford, so why should someone make a change and risk his job?

 

 

[Packer16]

Does anybody have a way to determine if Burford's underwriting criteria used to generate the great historical returns has not changed or will not change if alot of money floods this market?  In most credit situations you can look at credit metrics & trends in those metrics but what do you look at here?  I just do not see a moat here that will not be overcome if the alternative providers can provide the law firms a bigger piece of the pie.  It also appears that Burford's returns are skewed by Peterson but the company does not seem to point that out and lower expectations as a result which IMO could suggest aggressive assumptions in places we do not see.  From what I have read (Private Debt book by Nesbitt) the more normal returns in this business is in the low to mid teens.  The underwriting process appears opaque in terms of having an outsider get comfortable that it is repeatable with metrics they can understand.  Maybe I am missing something.  Let me know if I am.  TIA.

 

Packer

[cameronfen]

Does anybody have a way to determine if Burford's underwriting criteria used to generate the great historical returns has not changed or will not change if alot of money floods this market?  In most credit situations you can look at credit metrics & trends in those metrics but what do you look at here?  I just do not see a moat here that will not be overcome if the alternative providers can provide the law firms a bigger piece of the pie.  It also appears that Burford's returns are skewed by Peterson but the company does not seem to point that out and lower expectations as a result which IMO could suggest aggressive assumptions in places we do not see.  From what I have read (Private Debt book by Nesbitt) the more normal returns in this business is in the low to mid teens.  The underwriting process appears opaque in terms of having an outsider get comfortable that it is repeatable with metrics they can understand.  Maybe I am missing something.  Let me know if I am.  TIA.

 

Packer

 

Only on one point in your comment: I don't know what book you read, but almost all the players in the space that are public at least have IRRs in the 50+% range.  LIT, Manoette, IMF all report IRRs above 70%.

The reason those returns don't show up in RoE is because IRRs are before due diligence costs  (so just capital put in before capital returned).  The reason Burford has good RoEs is likely because they have the scale in terms of assets which allows them to spread personel costs.  Everyone realizes that so initially I think more capital flowing into this space will make players who take advantage of it more profitable, before it eats into IRRs I think.  But long term I agree Burford has no moat, and I don't think many people think it does.  The play is just can they compound capital fast enough to justify their market cap before returns go down.  I think they likely can but it's iffy although working in their favor is the fact that they are one of maybe 2 players attracting the lion share of capital flowing in. 

[cameronfen]

For anyone that's interested, since this thread was started, I've been working on IMF Bentham as a stock to buy as well as accumulating stock.  I wrote a pitch on it in it's own thread: http://www.cornerofberkshireandfairfax.ca/forum/investment-ideas/imf-asx-imf-bentham/

 

At first glance, IMF looks like the worst in breed business and trades at a comparably low 2.2x book value compared to comps.  On closer inspection though, management has probably the most effective and best long-term oriented strategy (which is why they are losing money right now), successfully executing a pivoting towards an asset management business model on track to post forward steady-state one year earnings of 50% of its entire market cap in a couple years down the road, which at the same time can grow exceptionally fast off that base at attractive ROIs.

[Pondside47]

Does anybody have a way to determine if Burford's underwriting criteria used to generate the great historical returns has not changed or will not change if alot of money floods this market?  In most credit situations you can look at credit metrics & trends in those metrics but what do you look at here?  I just do not see a moat here that will not be overcome if the alternative providers can provide the law firms a bigger piece of the pie.  It also appears that Burford's returns are skewed by Peterson but the company does not seem to point that out and lower expectations as a result which IMO could suggest aggressive assumptions in places we do not see.  From what I have read (Private  book by Nesbitt) the more normal returns in this business is in the low to mid teens.  The underwriting process appears opaque in terms of having an outsider get comfortable that it is repeatable with metrics they can understand.  Maybe I am missing something.  Let me know if I am.  TIA.

 

Packer

 

Only on one point in your comment: I don't know what book you read, but almost all the players in the space that are public at least have IRRs in the 50+% range.  LIT, Manoette, IMF all report IRRs above 70%.

The reason those returns don't show up in RoE is because IRRs are before due diligence costs  (so just capital put in before capital returned).  The reason Burford has good RoEs is likely because they have the scale in terms of assets which allows them to spread personel costs.  Everyone realizes that so initially I think more capital flowing into this space will make players who take advantage of it more profitable, before it eats into IRRs I think.  But  I agree Burford has no moat, and I don't think many people think it does.  The play is just can they compound capital fast enough to justify their market cap before returns go down.  I think they likely can but it's iffy although working in their favor is the fact that they are one of maybe 2 players attracting the lion share of capital flowing in.

 

In the medium term, this is an inefficient market. On the one hand you have law firms that are not set up structurally to finance these lawsuits. Generations of lawyers went through law school and training with the expectation of great financial rewards with little financial risk (Big law margins are 40-60%). It is going to take quite a while for them to change the mindset if ever. I'm not even sure frankly if they care to get a bigger slice of the profits Burford is capturing. Even people who are invested in Burford are debating if this is an industry that will generate great return in the  . Also lawyers are not MBAs. They are not trained to think in economic terms. It is possible a lot of them are just not attracted to the expected value of economic profit from uncertain activities such at litigation funding.

 

On the other hand, we have sophisticated players like Burford who is benefiting from a very inefficient market. We are very far away from securitizing these claims based on risk profile because to invest in these lawsuits you need a lot of confidential information. Right now the communication between a funder, law firm and the client is privileged. Once a claim is securitized, the communication probably won't remain confidential and that's a big problem for plenty law firm clients Petersen is an exception because there isn't much confidential information about the case anymore at this point.

 

So I'm quite confident Burford will remain profitable and will probably find even more ways to monetize their skills.  However, on the valuation front, considering the impact of Petersen, I'm not quite sure I have a large margin of safety now. The Petersen case can go south and that will have a big impact on the stock. Although I think it is likely the surprises will be on the upside.

 

Although even at ex Petersen P/E of 20, considering their invested asset base has more than doubled in the last three years, current price will be a P/E in the low teens in a couple of years. But again I can probably find more stocks with better moat that will "grow" into a reasonable P/E in 3 years.

 

 

[peterHK]

https://www.burfordcapital.com/wp-content/uploads/2019/07/2019.07.25-Burford-Capital-1H19-Interim-Report-WEB-FINAL-PUBLIC.pdf

 

H1 results are out and either they were as expected (revenue up, cash realizations up on Petersen sale etc.), and mgmt. Q&A in the report was exceptionally good and transparent including commentary on the company's AIM status (they'd rather uplist to NYSE/NASDAQ than LSE main board)

 

OR

 

they realized  a bunch of non cash gains and they're an accounting fraud and evil and terrible and the shares are down 6% because the market is figuring out they're a fraud and "stock price bro" proves it.

 

Which one you believe depends on whether you're already a bull or bear.

[Normax59]

https://www.burfordcapital.com/wp-content/uploads/2019/07/2019.07.25-Burford-Capital-1H19-Interim-Report-WEB-FINAL-PUBLIC.pdf

 

H1 results are out and either they were as expected (revenue up, cash realizations up on Petersen sale etc.), and mgmt. Q&A in the report was exceptionally good and transparent including commentary on the company's AIM status (they'd rather uplist to NYSE/NASDAQ than LSE main board)

 

OR

 

they realized  a bunch of non cash gains and they're an accounting fraud and evil and terrible and the shares are down 6% because the market is figuring out they're a fraud and "stock price bro" proves it.

 

Which one you believe depends on whether you're already a bull or bear.

 

It’s not black and white like that and no one here has even accused them of being fraudulent.

 

They reported FV gains as 50% of “income” once again, Petersen represented 98MM of realized gains with proceeds received down from last 1H. Not too sure how this could be viewed as a good report.

 

[Gregmal]

Earnings for a company like this will be lumpy and volatile. Marking illiquid or non realized gains to market is never perfect. Somehow hitting a few home runs is a bad thing? These concerns, and the subsequent stock decline just seem like stupid, fabricated, "Wall Street" issues.