BUR.L - Burford Capital

[Guest Schwab711]

You are right, it clearly says cumulative and I missed it. Stupid mistake on my part.

 

 

If anyone has the 2017 investment data pdf and would be willing to share it, I'd appreciate it.

[Jerry Capital]

It's on their website you just need to put in your email

[Guest Schwab711]

It's on their website you just need to put in your email

 

They give you 2018. i'm asking for prior years.

[Jerry Capital]

Burford case files attached

Burford-Capital-Investment-Data-2018.pdf

[Guest Schwab711]

I appreciate you attaching this. I was asking if anyone had this same file as of 12/31/2017 (or any other prior dates). Thank you for the effort though.

[Pondside47]

2018 is the first year they released the file.

[Jerry Capital]

My pleasure Schwab.

 

Curious, were you looking if there was potentially inconsistencies between the two? becuase the file they released this year has every case they have ever done.

 

Thanks

[Guest Schwab711]

My pleasure Schwab.

 

Curious, were you looking if there was potentially inconsistencies between the two? becuase the file they released this year has every case they have ever done.

 

Thanks

 

I would've checked for inconsistencies but I wouldn't have done it expecting to find any. After AZ_Value and VRX, I'm paranoid.

 

 

Assuming everything is fine as it has been, the main goal would be to recreate one year of revenue. With the 2017 table, we could track recoveries in 2018, movements from Ongoing -> Partially Realized or Concluded (and Partially Realized -> Concluded). We could look at all the tables BUR has provided to make guesses on various recognized balances. Depending how viable this idea was, we could even begin to make an estimate of how BUR estimates costs or what rulings led to what FV adjustment changes. That would probably take a lot of work and would probably be fruitless but anything to predict future PRs or financial results would be very valuable info.

 

BUR says we can't use this table to reconcile to the BS (which is fair) but it's a decent benchmark to figure out movements.

 

Outside of that, without knowing what the prior year looks like, I'm not sure what I'd find interesting so I'm not sure what I'd focus on. Maybe we could decipher what specific case some of these line items refer to and figure out how to search for these cases BUR is involved in preemptively in PACER. That might allow you to better predict future financial results or PRs. Maybe we can do this with a static look, I don't know.

[Pondside47]

If the figures I'm using are wrong feel free to correct me. Maybe I made a mistake somewhere. If you really think Jeffries stated what Petersen is on the books for can you show me a quote or something? It just contradicts what I know about fair value accounting so it's hard to believe as is. I can tell you how I looked at taxes but there's only so much we can read in book taxes, given the corporate structure. It looks like unrealized gains represents most of EBT in the last 3 years.

 

Just looking at the table on p.30 (2018 AR). Unrealized gains were 36% and 39% of investments in 2017 and 2018, respectively. That's $387m and $621m. That implies $234m of unrealized gains reported in 2018 revenue. That is pretty darn close to the calculated number for Petersen in 2018. If we repeat this for 2016 and 2017, we get $214m in unrealized gains reported as revenue in 2017. In 2016, unrealized gains represented $87m.

 

From the 2016 AR, Petersen was written up some, but no where near all the way to a $400m valuation. Perfect, we can see that. In 2017, BUR says they are aware of the $660m valuation trades but did not rely on them. If they went to $440m valuation at this point (the second direct sale at this price and conservative to secondary trading), that roughly fits what we measure (~$300m total unrealized gains + $106m proceeds on $440m valuation). If in 2018 they went to $800m (again, 3rd direct sale now and conservative to secondary trading), that again fits what we can calculate.

 

All of this fit the situation of negative book taxes with high reported income. Taxable income would be much lower with these unrealized gains. This all fits the cash flow statements. If there's issues with breaking out Petersen like this I'm open to hearing them but this seems all seems consistent with everything I've read from BUR. That means Petersen represents the majority of earnings between 2016-2018.

 

As to the quality of the business, I agree with you and others that it's a nice business. I may disagree on how to value it but it's a business I'd like to own. I like the trends in lit finance, I like that cash flows are generally uncorrelated, and so on. Packer brought up a great point about PFIC that I can't find a specific answer to, but it sure seems like it would be classified as such. I don't know much about all of that so that's a bit concerning. I don't know what I'd pay for BUR yet but I think it's worth getting comfortable with them since I would own them at some price.

 

So if you look at the Burford Capital Investment Data (as of December 2018) that was linked by Jerry Capital before, you can actually find the Petersen case. The Teinver case is 2010 Partially realized Single Case 111065 $13mm deployed $107 recovered. The Petersen case is 2015 partially Realized Portfolio 143539 Contract Energy NA $18.2 deployed, $136 Recovered. You could argue what's shown in the spread sheet is not what's used to calculate the revenue, it's possible.

 

On a negative note, I found several significant typos in the spreadsheet such as 2013 Concluded Portolio 178742 Antitrust Mixed NA $3.9 deployed $10.3 recovered, but the ROIC is shown as 346%. Also I couldn't get to the advertised 41% ROIC and 15% IRR (Page 21 2018 AR) for the period prior to 2016 if you write down everything ongoing (very aggressive assumption since ongoing deployed is 50% of the total deployed) to 0  from the numbers in the PDF file. However, I think what's shown in the PDF file is quite conservative.

 

[Normax59]

If the figures I'm using are wrong feel free to correct me. Maybe I made a mistake somewhere. If you really think Jeffries stated what Petersen is on the books for can you show me a quote or something? It just contradicts what I know about fair value accounting so it's hard to believe as is. I can tell you how I looked at taxes but there's only so much we can read in book taxes, given the corporate structure. It looks like unrealized gains represents most of EBT in the last 3 years.

 

Just looking at the table on p.30 (2018 AR). Unrealized gains were 36% and 39% of investments in 2017 and 2018, respectively. That's $387m and $621m. That implies $234m of unrealized gains reported in 2018 revenue. That is pretty darn close to the calculated number for Petersen in 2018. If we repeat this for 2016 and 2017, we get $214m in unrealized gains reported as revenue in 2017. In 2016, unrealized gains represented $87m.

 

From the 2016 AR, Petersen was written up some, but no where near all the way to a $400m valuation. Perfect, we can see that. In 2017, BUR says they are aware of the $660m valuation trades but did not rely on them. If they went to $440m valuation at this point (the second direct sale at this price and conservative to secondary trading), that roughly fits what we measure (~$300m total unrealized gains + $106m proceeds on $440m valuation). If in 2018 they went to $800m (again, 3rd direct sale now and conservative to secondary trading), that again fits what we can calculate.

 

All of this fit the situation of negative book taxes with high reported income. Taxable income would be much lower with these unrealized gains. This all fits the cash flow statements. If there's issues with breaking out Petersen like this I'm open to hearing them but this seems all seems consistent with everything I've read from BUR. That means Petersen represents the majority of earnings between 2016-2018.

 

As to the quality of the business, I agree with you and others that it's a nice business. I may disagree on how to value it but it's a business I'd like to own. I like the trends in lit finance, I like that cash flows are generally uncorrelated, and so on. Packer brought up a great point about PFIC that I can't find a specific answer to, but it sure seems like it would be classified as such. I don't know much about all of that so that's a bit concerning. I don't know what I'd pay for BUR yet but I think it's worth getting comfortable with them since I would own them at some price.

 

So if you look at the Burford Capital Investment Data (as of December 2018) that was linked by Jerry Capital before, you can actually find the Petersen case. The Teinver case is 2010 Partially realized Single Case 111065 $13mm deployed $107 recovered. The Petersen case is 2015 partially Realized Portfolio 143539 Contract Energy NA $18.2 deployed, $136 Recovered. You could argue what's shown in the spread sheet is not what's used to calculate the revenue, it's possible.

 

On a negative note, I found several significant typos in the spreadsheet such as 2013 Concluded Portolio 178742 Antitrust Mixed NA $3.9 deployed $10.3 recovered, but the ROIC is shown as 346%. Also I couldn't get to the advertised 41% ROIC and 15% IRR (Page 21 2018 AR) for the period prior to 2016 if you write down everything ongoing (very aggressive assumption since ongoing deployed is 50% of the total deployed) to 0  from the numbers in the PDF file. However, I think what's shown in the PDF file is quite conservative.

 

It's explained in the document that it's not reconcilable to the annual report.

[Normax59]

My pleasure Schwab.

 

Curious, were you looking if there was potentially inconsistencies between the two? becuase the file they released this year has every case they have ever done.

 

Thanks

 

I would've checked for inconsistencies but I wouldn't have done it expecting to find any. After AZ_Value and VRX, I'm paranoid.

 

 

Assuming everything is fine as it has been, the main goal would be to recreate one year of revenue. With the 2017 table, we could track recoveries in 2018, movements from Ongoing -> Partially Realized or Concluded (and Partially Realized -> Concluded). We could look at all the tables BUR has provided to make guesses on various recognized balances. Depending how viable this idea was, we could even begin to make an estimate of how BUR estimates costs or what rulings led to what FV adjustment changes. That would probably take a lot of work and would probably be fruitless but anything to predict future PRs or financial results would be very valuable info.

 

BUR says we can't use this table to reconcile to the BS (which is fair) but it's a decent benchmark to figure out movements.

 

Outside of that, without knowing what the prior year looks like, I'm not sure what I'd find interesting so I'm not sure what I'd focus on. Maybe we could decipher what specific case some of these line items refer to and figure out how to search for these cases BUR is involved in preemptively in PACER. That might allow you to better predict future financial results or PRs. Maybe we can do this with a static look, I don't know.

 

Because they give the weighted average duration along with the IRR it's possible to reverse engineer what years the cash flows came in without reports from other years, though it would take quite a while to do each individual deal. Not sure if its even worth it though as you said.

[peterHK]

Maybe I'm not paranoid enough, but it seems to me that this is best treated as an investment fund. One can't forensically audit absolutely everything, and IMO the time spent on doing such falls in the 80% of the work delivering 20% of the value.

 

Time is far better spent here looking at management and figuring out the character of who is running Burford at a variety of levels from compliance to the CEO/CFO/CIO levels.

[Normax59]

Maybe I'm not paranoid enough, but it seems to me that this is best treated as an investment fund. One can't forensically audit absolutely everything, and IMO the time spent on doing such falls in the 80% of the work delivering 20% of the value.

 

Time is far better spent here looking at management and figuring out the character of who is running Burford at a variety of levels from compliance to the CEO/CFO/CIO levels.

 

I'd rather just look at the numbers and figure out if they are accurate and represent the reality of the business, since they gave all the info that would be needed to track the cash flows from recoveries from the document, over not doing the work and crossing my fingers and hoping management is ethical.

 

Currently converting the PDF to excel and replicating the cash flows by year based on the IRR's and weighted duration's that they gave, I'll post it whenever its finished for those that would want to toy with it.

[Jerry Capital]

FCF less than earnings is usually a bad thing.

 

"Low Quality Earnings"

 

If you understand the company and the flow of funds it makes perfect sense that cash doesn't track earnings. In fact they could juice near-term cash flow by selling all of Peterson (in which case FCF would be greater than NI), but they want to realize the higher cash payout I a few years down the road because they think it's worth more than the current secondary market.

 

I Beleive they would sell the entire case if the secondary market discounted at a 20 to 30% cost of capital, which is probabaly their hurdle rate...

 

This lowers the compounding effect and reduces the companies ability to gear its balance sheet.

[Normax59]

I don't know of any other company that operates in the FIG sector with 77% after tax margins, but maybe I'm not looking hard enough. The whole thing screams to good to be true.

[Jerry Capital]

There is usually a lot of alpha in situations where you can't quite "box" it into a typical sector, industry, coverage list.

 

This is something new and evolving. If you are right and your implication is... This is a fraud? Or you are implying what they are doing is not sustainable? (don't want to put words in your mouth!

 

If you have more evidence than your hunch I would be interested.

 

For what it is worth I tend to have that same feeling, that it's "smelly".

 

I am pushing against that "intuition" because I think I can see downside protection not to far away, with a really great story.

 

[Normax59]

There is usually a lot of alpha in situations where you can't quite "box" it into a typical sector, industry, coverage list.

 

This is something new and evolving. If you are right and your implication is... This is a fraud? Or you are implying what they are doing is not sustainable? (don't want to put words in your mouth!

 

If you have more evidence than your hunch I would be interested.

 

For what it is worth I tend to have that same feeling, that it's "smelly".

 

I am pushing against that "intuition" because I think I can see downside protection not to far away, with a really great story.

 

I'm implying that its not sustainable. If I thought it was a fraud I would have said that and provided evidence, that is a big accusation to lob.

[Jerry Capital]

Of course.

 

I just again, I don't want to put words in your mouth.

 

May I ask what specifically do you think is not sustainable here?

[Guest Schwab711]

Petersen.

 

This entire investment is Petersen, the current accounting of it and the frequency of cases of its magnitude. As BUR grows, they will soon need a Petersen every year to support the current valuation. If BUR grows larger based on speculated TAM, they'll need to average more than one Petersen a year.

 

If we assume Petersen is a one-off that cannot be repeated in magnitude then BUR's LT ROE probably maxes out at 15%-ish and true LT ROE is probably something less than 12%. That supports a much smaller P/BV than 3+.

 

 

Ex-Petersen, BUR's cumulative ROIC is ~1.6x on completions and ROIC has been falling over time. Obviously settlements have lower ROIC than trial cases, so near term should be lower. With that said, ex-Petersen, ROICs for 2015-2017 are 14%, 27%, and 26%. If that covers overhead (~8% annually at present) it's not by much considering the average life of a claim.

 

Two cases account for ~50% of cumulative recoveries for BUR. I get concerns about eliminating the best cases but it only points to the fact that we should expect lumpy financial results. The smooth 30% returns are not indicative of the future, even if you expect LT 30% ROE.

 

Edit: I think an unmentioned risk is future claims may have capped gains. If Petersen-like outcomes (relative and absolute return) are expected to be somewhat frequent then law firms or clients may argue that claims be capped at 10x return or so (as an arbitrary example). Outside negotiating leverage to extract a cap, various states are discussing regulating a cap on contingent returns as they consider litigation financing to be similar to payday loans.

 

See p.13/24

https://www.aciclaw.org/system/files/events/attachments/2017/2017-12-07_acic_presentation_litigation_finance.pdf

 

https://hireanesquire.com/blog/2018/8/27/legal-marketing-trend-litigation-portfolio-funding

[Normax59]

Petersen.

 

This entire investment is Petersen, the current accounting of it and the frequency of cases of its magnitude. As BUR grows, they will soon need a Petersen every year to support the current valuation. If BUR grows as larger based on speculated TAM, they'll need to average more than one Petersen a year.

 

If we assume Petersen is a one-off that cannot be repeated in magnitude then BUR's LT ROE probably maxes out at 15%-ish and true LT ROE is probably something less than 12%. That supports a much smaller P/BV than 3+.

 

 

Ex-Petersen, BUR's cumulative ROIC is ~1.6x on completions and ROIC has been falling over time. Obviously settlements have lower ROIC than trial cases, so near term should be lower. With that said, ex-Petersen, ROICs for 2015-2017 are 14%, 27%, and 26%. If that covers overhead (~8% annually at present) it's not by much considering the average life of a claim.

 

Two cases account for ~50% of cumulative recoveries for BUR. I get concerns about eliminating the best cases but it only points to the fact that we should expect lumpy financial results. The smooth 30% returns are not indicative of the future, even if you expect LT 30% ROE.

 

I was going to point my finger at Petersen and unrealized gains as well, but you could even take a top-down approach we can assume they find a Petersen every year. I'm of the opinion that banking/lending/finance is a commodity business with very little secrets, if this is scale-able with the margins they have nothing stops those with big pockets from just throwing money at it, its not like they couldn't find the best lawyers in the world. Though an acquisition argument could be made.

[Happy]

So far this has been a high ROE, strong growth business that can reinvest all its earnings at high rates and even raises additional capital like the sovereign wealth fund because they claim they find more good opportunities that they can invest in on their own. The management seems to have done great moves so far like their acquisitions, which quickly paid for themselves (and then some). I feel like this is priced as if the returns on capital will deteriorate rapidly and the growth will come way down. I can't say that this is impossible, but it seems far from a sure thing and if they can keep returns up for longer than many think and/or just keep growing as they have by entering Australia, Asia, Germany etc, then it could prove to still be very cheap here. Probabilistically Iike the risk/reward here.

 

I'm sure people thought that the returns of Constellation Software or Credit Acceptance were too good to be true or sustainable as well and yet they just kept going. They don't have uncrossable moats either, but some advantages and great execution/management. What gives me some comfort that their numbers are real and that this might go on longer is that the big US law firms have had around 40% net profit margins for a long time and other litigation finance players show even higher IRRs than Burford. Law seems to be a very lucrative field in general, like software etc.

 

If the price goes significantly higher with no news, then it becomes a tough decision how long they are likely to keep the returns up etc. But at this price it seems like you are paying a price that assumes they go down fast and all scenarios where they go down much slower or they keep growing for a long time are pure upside.

 

 

[Pondside47]

Petersen.

 

This entire investment is Petersen, the current accounting of it and the frequency of cases of its magnitude. As BUR grows, they will soon need a Petersen every year to support the current valuation. If BUR grows as larger based on speculated TAM, they'll need to average more than one Petersen a year.

 

If we assume Petersen is a one-off that cannot be repeated in magnitude then BUR's LT ROE probably maxes out at 15%-ish and true LT ROE is probably something less than 12%. That supports a much smaller P/BV than 3+.

 

 

Ex-Petersen, BUR's cumulative ROIC is ~1.6x on completions and ROIC has been falling over time. Obviously settlements have lower ROIC than trial cases, so near term should be lower. With that said, ex-Petersen, ROICs for 2015-2017 are 14%, 27%, and 26%. If that covers overhead (~8% annually at present) it's not by much considering the average life of a claim.

 

Two cases account for ~50% of cumulative recoveries for BUR. I get concerns about eliminating the best cases but it only points to the fact that we should expect lumpy financial results. The smooth 30% returns are not indicative of the future, even if you expect LT 30% ROE.

 

I was going to point my finger at Petersen and unrealized gains as well, but you could even take a top-down approach we can assume they find a Petersen every year. I'm of the opinion that banking/lending/finance is a commodity business with very little secrets, if this is scale-able with the margins they have nothing stops those with big pockets from just throwing money at it, its not like they couldn't find the best lawyers in the world. Though an acquisition argument could be made.

 

I don’t think this is banking or lending where assets are homogeneous. I have a hard time comparing this to reinsurance either at least not property reinsurance. If this is a 10-12% roe no growth business in the long term but the earnings are uncorrelated to the market and they have a good track record, I think this will trade at a much higher multiple than 1.5.

 

 

[Jerry Capital]

It is impossible to predict if they will have another Peterson, it is unlikely for sure. Again it is similar to a VC fund, I would invest in the next round of Gurleys investments and he is operating in the frothiest of markets (see Softbank), yet I believe he will continue to generate significant alpha over time. But I think its important to point out that EVEN if funds came into litigation finance like they are now, if that accelerated and it became completed saturated, there is still a probability that they underwrite good investments.

 

That said, I believe it is more likely that the industry will remain structurally inefficient for a vareity of reasons that will continue for the foreseeable future. The leader in a structurally inefficiient market certainly will take their share of 1) fund flows, 2) attractive investments.

 

I also think that it is important to note that management is trying to shift this business away from "needing another peterson" and more towards a company that should be valued on actual cash flows, the asset management arm, the portfolio financing (lower ROIC but revenues and cash flows become more stable overtime = multiple of FCF not multiple of BV). This is the right thing to be doing. We will probably look back in a couple of years at they will not have completed this transition, the stock will have underperformed, maybe even a permenant loss of capital. But I have a hard time seeing it being much more than 30% from $15 per share (50%? OK size your position accordingly, this is not supposed to be in your portfolio at 20% weight!). Even CG has a $12.5 price target (or close to that) and they pretty much laid out the most bearish case possible on the company.

 

The other side of the coin, albeit a lower probability, is a lot of things work out for them, and alot of things work out for them when the S&P 500 goes through a bear market, the economy goes through a recession, etc. If Burford can show that they are recession resistant, that they transition to a FCF valued firm with semi recurring revenues, the jsutified mutliple will be insanely high, you will make A LOT of money.

 

At the right size, this is an interesting stock to put in your portfolio to get convexity and a lot of potential alpha should we hit a ugly bear market.

[Normax59]

So far this has been a high ROE, strong growth business that can reinvest all its earnings at high rates and even raises additional capital like the sovereign wealth fund because they claim they find more good opportunities that they can invest in on their own. The management seems to have done great moves so far like their acquisitions, which quickly paid for themselves (and then some). I feel like this is priced as if the returns on capital will deteriorate rapidly and the growth will come way down. I can't say that this is impossible, but it seems far from a sure thing and if they can keep returns up for longer than many think and/or just keep growing as they have by entering Australia, Asia, Germany etc, then it could prove to still be very cheap here. Probabilistically Iike the risk/reward here.

 

I'm sure people thought that the returns of Constellation Software or Credit Acceptance were too good to be true or sustainable as well and yet they just kept going. They don't have uncrossable moats either, but some advantages and great execution/management. What gives me some comfort that their numbers are real and that this might go on longer is that the big US law firms have had around 40% net profit margins for a long time and other litigation finance players show even higher IRRs than Burford. Law seems to be a very lucrative field in general, like software etc.

 

If the price goes significantly higher with no news, then it becomes a tough decision how long they are likely to keep the returns up etc. But at this price it seems like you are paying a price that assumes they go down fast and all scenarios where they go down much slower or they keep growing for a long time are pure upside.

 

The two companies you mentioned didn't have to rely on a single customer/event to provide them their out sized returns, neither had 75-85% of their before taxes earning being made up of unrealized gains, but I understand the point.

[cameronfen]

Petersen.

 

This entire investment is Petersen, the current accounting of it and the frequency of cases of its magnitude. As BUR grows, they will soon need a Petersen every year to support the current valuation. If BUR grows as larger based on speculated TAM, they'll need to average more than one Petersen a year.

 

If we assume Petersen is a one-off that cannot be repeated in magnitude then BUR's LT ROE probably maxes out at 15%-ish and true LT ROE is probably something less than 12%. That supports a much smaller P/BV than 3+.

 

 

Ex-Petersen, BUR's cumulative ROIC is ~1.6x on completions and ROIC has been falling over time. Obviously settlements have lower ROIC than trial cases, so near term should be lower. With that said, ex-Petersen, ROICs for 2015-2017 are 14%, 27%, and 26%. If that covers overhead (~8% annually at present) it's not by much considering the average life of a claim.

 

Two cases account for ~50% of cumulative recoveries for BUR. I get concerns about eliminating the best cases but it only points to the fact that we should expect lumpy financial results. The smooth 30% returns are not indicative of the future, even if you expect LT 30% ROE.

 

I was going to point my finger at Petersen and unrealized gains as well, but you could even take a top-down approach we can assume they find a Petersen every year. I'm of the opinion that banking/lending/finance is a commodity business with very little secrets, if this is scale-able with the margins they have nothing stops those with big pockets from just throwing money at it, its not like they couldn't find the best lawyers in the world. Though an acquisition argument could be made.

 

I don’t think this is banking or lending where assets are homogeneous. I have a hard time comparing this to reinsurance either at least not property reinsurance. If this is a 10-12% roe no growth business in the long term but the earnings are uncorrelated to the market and they have a good track record, I think this will trade at a much higher multiple than 1.5.

 

So maybe something to compare this to is catastrophe bonds.  Cat bonds used to be a high returning asset that was uncorrelated with the markets and then a lot of capital flowed in and returns began to normalize.  It might be interesting to have a guidepost about at what % of TAM did ROEs begin to normalize.