18% growth in BPS is pretty good, but they also grew BPS at a +16.5%/year clip over five years, +13.0%/year over ten years and +15.3%/year over 20 years. Those are some pretty impressive figures.
Out of curiousity, I put all of this into the table below. The first three columns come right out of the BRK annual report. I added the five columns to the right from MKL's annual reports. MKL has done better than BRK in terms of BPS growth in five, ten and twenty year time periods.
Sometimes questions come up about MKL's equity portfolio; is it really that good? So I added the last two columns which shows the return of MKL's equity portfolio over time. This shows some serious outperformance against the total return of the S&P 500 index over five, ten and twenty year periods.
What is interesting is that not only did they outperform in the longer time periods, but they outperformed in each of 2010, 2011, 2012 and 2013; years that were very strong in the stock market which might cause more conservative portfolios to underperform.
BRK, MKL and MKL's Equity Portfolio versus the S&P 500 Index
*Five and ten year equity portfolio returns don't match figures in table; I used the number reported in the 2013 annual report. Compounding the figures in the above table would give you +20.2%/year and +9.1%/year for the five and ten year returns.
Five Year Test
Using Buffett's five year test, which is actually no longer Buffett's five year test, MKL has failed to outdo the S&P 500 index over the past five years, just like BRK. So let's look at Buffett's "through the cycle" test.
Performance from 2007 peak through 2013 (and 1999-2013)
Buffett said in the BRK 2013 annual report that:
Over the stock market cycle between yearends 2007 and 2013, we outperformed the S&P 500 index. Through full cycles in future years, we expect to do that again. If we fail to do so, we will not have earned our pay. After all, you could always own an index fund an be assured of S&P results.But why stop there? Since I'm doing the work, I might as well toss in the annualized returns for all of this since the 1999 peak so we get two whole cycles.
Here is what the above figures look like from December 2007 - December 2013 and December 1999 through December 2013:
S&P 500 Index +6.3%/year +3.6%/year
BRK +9.6%/year +9.5%/year
MKL +10.3%/year +14.9%/year
MKL Equity portfolio + 8.8%/year +10.8%/year
MKL's BPS grew 9.6%/year versus the S&P's +6.3%/year through this cycle; that's 3.3%/year better. The equity portfolio also outperformed since the 2007 peak.
The comparison is even better if you look at it since 1999; MKL's BPS outperformed the S&P 500 index by 11.3%/year! That's better than Buffett's goals during his partnership years (10% better than the index). The equity portfolio also outperformed the index by 7.2%/year. That's pretty big.
OK, so Markel had a really good year. They closed the Alterra acquisition, their combined ratios look good, businesses are growing etc. I won't get into all of that here so I'll just paste a short summary from the report:
Portfolio Per Share
So here's a metric that MKL usually shows. Portfolio per share grew to $1,259/share in 2013, versus the current stock price of $591/share. Why is this important? Check out what MKL has to say about this:
One common way to value BRK is the two column method. That is to calculate the investments per share of BRK and add it to some multiple of the profits earned by the wholly owned businesses (businesses not in the investment porfolio). This method treats insurance "float" as equity as they don't have to be paid back and BRK typically, over time, earns a profit on insurance underwriting (cost of float is negative).
MKL too has a pretty good underwriting record and has earned a profit over time. Is their "float", then, not equity? If it is, then MKL's intrinsic value, using the same approach as BRK's two column method, would be portfolio per share (less debt).
But I don't raise this issue to debate the two column method or whether float is equity. It's just a thought. Hmm...
Shift in Emphasis
Here's a new, interesting thing in the annual report. From now on, they want us to focus on BPS CAGR (compound annual growth rate) instead of BPS itself. Here is their explanation:
And they elaborate later in the report (I just cut some stuff out, this is not the whole section so please read the whole letter):
BPS-CAGR to Earnings and then to P/E
So obviously, a company that is growing book at a faster pace than others should be more valuable. This is obvious. But how much higher should the valuation be? There are many ways to do this, of course. One thing people would do when looking at financials is to use a 10% hurdle; if a company grows book at 8%, it's worth 0.8x book. If it grows at 12%/year, it's worth 1.2x book. Never mind if that makes any sense or not, but that's a quick rule of thumb thing that people sometimes look at.
Since MKL wants us to focus on the earnings/flow aspect of MKL instead of the stock (BPS), why not convert this BPS-CAGR into a sort of earnings figure? Change in BPS, simplified, is basically comprehensive income per share; kind of like an all inclusive EPS.
So if CAGR of BPS over time is 15%, then MKL comprehensive EPS is like $72/share. At 10x this figure, MKL is worth $720/share. But if BPS CAGR is 15%/year, so is comprehensive EPS growth.
If the $72/share EPS is going to grow 15%/year, what multiple would you put on it? At a very conservative 15x, MKL would be worth $1,080/share. If MKL can really grow 15%/year, then maybe it's worth 20x p/e. That would put MKL at $1,440/share.
Remember, MKL has grown BPS at 16.5%/year over five years, +13.0%/year over ten years and +15.3%/year over twenty years.
So let's use 13%, then. Comprehensive EPS would be $62, and this would grow 13%/year. Put a 13x multiple on it (don't ask me, but growth investors like to pay the same p/e as the growth rate. Even Julian Robertson did that in an interview a while back). That would put MKL's value at $806/share. This sort of makes no sense since a 13% growth rate is way better than the S&P 500 index, right? So let's say p/e of 15-20x is more reasonable. That's $930 - $1,240/share in value for MKL.
OK, so I am not going to tell you that MKL is worth $1,440/share, or any of the above values. I'm just playing around with things to see what would happen.
I think the market would still view MKL as primarily a financial company (so will keep emphasizing BPS for a while, maybe until MKL Ventures gets much bigger); much of their book is still marked-to-market, and that will anchor valuations to BPS somewhat. But maybe less and less over time.
Oh, and I almost forgot to mention this. MKL said that after the Alterra acquisition, their equity portfolio as a percentage of shareholders' equity went below 50%, but after building up their equity portfolio throughout 2013, it is back up to 49%. And they said of this ratio that "we would expect it to gradually increase towards our more normal target of 80% over time".
This is very exciting. Think about it. To simplify, let's say that this ratio gets up to 100%. That means that you get an equity return (better than the index if history keeps rhyming) on the total book value and everything else comes as a bonus; if interest rates go up, then the interest income is additive to this. If underwriting makes a profit, that's a bonus too. And if Markel Ventures keeps growing, whatever you gain on that would be on top of the equity portfolio returns. What's not to like? (OK, well, you would have to deduct interest expense on debt. And things can always get ugly with some 'engines' going into reverse).
Anyway, MKL has grown book at a much higher pace than I thought they could in this environment of not-so-great insurance market and low interest rates. If they can keep this up, 1.2x book looks like a pretty decent bargain!
Thanks, was looking forward to the MKL letter.ReplyDelete
I think the market would still view MKL as primarily a financial company (so will keep emphasizing BPS for a while, maybe until MKL Ventures gets much bigger)
Then when MKL Ventures gets bigger, the market will slap a conglomerate/black box discount? =)
Yes, of course. Then people will demand a dividend, stock split, or a complete spin-off of something.Delete
I really don't think it is prudent to model that MKL is going to grow at 15% for a really long time or any company for that matter unless they are tech based and have ridiculous scaling abilities. Where's the margin of safety if you have to assume Pollyanna forecasts? Assuming half the historical growth to figure out intrinsic value will usually lead to better results.ReplyDelete
The margin of safety with Markel at the current price is that we aren't paying for any of those assumptions. The market views Markel as a boring financial, not at a company capable of replicating its historical record. But here's the thing: Even if the P/B NEVER expands, if MKL can compound book value at 12% or 15% or 18% per year, then that's what stockholders are likely to earn over time. In other words, at the end of the next decade or two, the market can still insist that Markel isn't a growth story and give it a P/B similar to 1.2 or 1.3 and shareholders still win. Ultimately, value is reflected in the price one way or another. P/B expansion is just gravy.Delete
I agree that modelling 15%/year for a long time is usually not prudent. But MKL has done that for a very long time. You can use other numbers too for sure. I used various numbers as examples and you can do the same analysis with different growth assumptions and p/e ratios. You can say 10-12%/year BPS growth and a 15x p/e, for example.Delete
And yes, it's not aggressive at all to assume P/B can be 1.2 or 1.3x over time. (second poster).
That means that you get an equity return (better than the index if history keeps rhyming) on the total book value and everything else comes as a bonus; if interest rates go up, then the interest income is additive to this. If underwriting makes a profit, that's a bonus too.ReplyDelete
There is also that issue of double taxation...
Yes, double taxation is an issue. But if MKL is a long term investor, owed taxes become an interest free loan and source of book value leverage (like BRK; much of BRK's investment/book value leverage comes from tax liabilities). Also on the MKL shareholder end, if they are long term holders, long term gains payable can be deferred for a long time.Delete
So net, net, you are probably still way better off than owning a mutual fund with 100% turnover that merely matches the index (you'd be lucky if you did match index). Thinking of this in isolation, it seems like an index fund, then, might be a better idea, but the above figures show, despite the tax headwind, that MKL has been a better investment than an index.
Great stuff as usual...ReplyDelete
Great biz model (Berkshire), conservative top class mgmt, float and and Gayner. My top position by far and one I have full intention of owning for decades. And selling for reasonably cheap Price to Book doesn't hurt.
Do you follow ORI at all? Book value has grown at more than 15% for 50 years. Although they have had a tough go of it for the last 5 years due to problems in their mortgage guaranty line. And do you ever look at Return on Equity drivers, which is a modified DuPont analysis for insurance companies? It is a nice breakout of how much ROE comes from underwriting profit vs investments.ReplyDelete
No, I don't follow ORI but I'll take a look. I do look at ROE drivers even though I never explicitly calculate the DuPont analysis exactly the way they do. But I do look at return on investment portfolio, investment leverage, underwriting profit/loss and how that drives ROE, so I guess I do in way.Delete
Nice graph showing Alterra vs. Peers
Thanks, kk. Yep, wide bracket on MKL value, but I got pretty close to that $1,440 bogey you mentioned, too, in two different hacks using Gayner and Berkowitz's own approaches. More here FWIW: https://www.islainvest.com/wp-content/uploads/2013/10/VIC-MKL-Submission-Final-v3.pdfReplyDelete
The bruce berkowitz's method seems to be misleading lol. U gotta discount the 1438 per share by whatever discount rate/hurdle rate you are expecting for markel. If you discount it back by 10% u get back 500 ish.Delete
Also, a comment on alice schroeder method, does anyone think that the schroeder float-based method (i assumed that its the same as the 1999 paine webber brk method) seems to theoretically incorrect? It seems to ignore all the first 10 years float's returns (which could be used to repurchase shares in the case of MKL since they don't pay dividends). This would actually boost the intrinsic value of MKL. (pls correct me if i'm wrong, not sure if i'm getting this valuation method correctly or not).
Thanks for this investment thesis. Looking at the tricky part of valuation approaches, it comes down to the float. Equity or debt? Since it’s essentially money owed and never paid back. That said, looking at -say- the 2005 financials, float amounted to $ 5.9 bn vs $ 5.4 bn end of 2012. Thus, on an organic basis, MKL has used less and less float, which could be treated as a debt repayment, right? Treating it as a pure debt, which is not conceptually right, would seriously trim down #1 and #4 approaches. Regarding #3 approach, why don’t you discount back the $1,438 figure to the present? Growth rate in the #2 approach seems fairly optimistic, but here anyone can plug in their estimates.Delete
I'm not trying to get an exact number but a sense if MKL is cheap enough for a decent margin of safety.
Yeah, but MKL gets there by using 70% more leverage than BRK. Works fine until it doesn't...ReplyDelete
Well, MKL has been pretty well stress-tested over the past 20 years or more... They've gone through the financial crisis (no TARP money there!) and some pretty historically bad insurance years too that wiped out many insurance companies.Delete
So yes, works fine until it doesn't, I suppose, but they've been through many cycles already...
If you go through their shareholder letters you will find some references to the mid 90s. The market was soft during that time with lots of new players entering (similar to what we see today in the reinsurance market). A few major events later the new players were flushed out and MKL happily picked up the remainders. They also did quite well after 9/11, Sandy, Katharina and friends and the like. So I would say that they know what they are doing.Delete
KK, have you thought about a potential acquisition of Berkshire here? Not sure if they would be complementary or not but would like your thoughts. We hear them compared to BRK pretty often around the message boards, etc. Would also potentially add to the stable of high quality investors, especially given the records you discuss above.ReplyDelete
Thanks again for all of this. You have been on fire lately with your postings.
Acquisition by Berkshire sorry not acquisition of Berkshire lolReplyDelete
Yes, that's an interesting thought and I've heard others mention it. You know, anything is possible, but MKL is actually doing pretty well now so I would guess that they like their independence and creating their own 'masterpiece' if you will. So although they would be a great fit culturally etc... I don't know that there is any motivation on either side at this point... but who knows.
I’ve recently become very interested in investing over the past 2-3 years and I’ve found your blog to be educational, interesting, and first class. Thank you and I look forward to continuing to read your posts.ReplyDelete
One thing I noticed is your post is titled “Markel 2013 Annual Report” however you focus mostly on the content in the letter to shareholders. I was wondering if you noticed the additional information they provided this year regarding Markel Ventures (Balance Sheets, Earnings, and Cash Flows compared to just EBITDA and Earnings in the past).
My calculations showed that Markel Ventures earned about 10% after tax (excluding purchase-accounting items) on capital employed. After reading your series on Buffett and market timing, it appears Markel is getting even better than the 10% pre-tax returns Buffett has stated as a hurdle.
Additionally Markel Ventures is earning about 18% after tax (excluding purchase-accounting items) on tangible equity + debt. For comparisons sake, Berkshire’s Manufacturing, Service and Retail operations are collectively earning 16.7% after tax on tangible equity with basically no debt.
Overall, I thought these were satisfactory and demonstrated that Gayner and Markel have successfully used their experience in equities to make attractive private deals. I was wondering if you looked at these additional disclosures and had any thoughts on Markel Ventures.
Yes, that's a good point. I often title things "annual report" and actually just talk about the letter to shareholders. I do look at everything else too, but these annual report posts are not meant to be detailed look at the financials. I will only point something out or bring it up as a topic if there is anything of interest specifically that I want to talk about.
Yes, MKL Ventures seems to be taking off, but I haven't really dug into it much as there really isn't much to dig into yet. It's still early in the game.
As for 10% pretax hurdle, yes, MKL achieves that no problem on it's own capital, and buyers of the stock now can get that too.
Thanks for reading.
Is anyone worried about succession here? Chairman/CEO Kirshner is 78. How much of this performance and culture is due to him?ReplyDelete
I found a good article about him from 2006
Kirshner has been running things for a long time:
"A major turning point for Kirshner was in the late 1970s, when the senior Markels placed their common stock with the third generation, Steve and Tony, and Kirshner. The elder Markels were making the statement that the business wouldn't be broken up or sold."
Seems like he has a strong personality:
"If something's going wrong, you want to call and get somebody in charge. So if we got nine presidents, plus a corporate president, you ought to get a damn president if you want to."
That's a good question. I haven't really worried about it as it seems like the folks there have really built up the culture together and it's not like BRK where Buffett makes the key investment and acquisition decisions.Delete
Thanks for the post. As far as valuation is concerned, here is what T. Gayner said early 2012 : "I would not put a specific number on it, but the mentality that I would have in thinking about it is, I would think about a our insurance businesses with a normalized amount of premium volume, and a normalized amount of underlying profitability. That gives a normalized amount of operating earnings from the insurance company. Then I would look at our net investments per share, which was total investments minus all the debt. I would then think to myself, well, if this insurance company continues to operate at an underwriting profit, and at least stays the same size, then all of the net investments per share are actually working for the shareholder, so that value should be added. Then the third thing I would think about is the Markel ventures set of companies and what the operating cash flows and operating earnings are on a normalized basis and assign a multiple of that. I would divide that by the number of shares outstanding, and get a sense of what Markel each share is worth."ReplyDelete
Have you had a chance to look at WR Berkley ? WRB’s insurance business (niche markets) seems quite close to MKL’s. Over the last 10 years, WRB has compounded their BPS a little faster, +14.7% dividends included, than MKL. Yet, the path to those results is quite different: a combined ratio a touch better at WRB although I’m not sure whether their reserves’ policy is as prudent as MKL’s, more fixed-income securities and leverage at WRB, more mergers or acquisitions at MKL. Both companies trade around 1.3x BV.
Hi, yes, I have looked at WRB and it's great operation. I tend to be focused, though, on the guys that do both active investing and good underwriting. But WRB is good too.Delete
Thanks for the post. Where did you find the historical returns on the Markel equity portfolio? Why do you think these do not match the 5yr and 10yr figures management has given?ReplyDelete
The annual reports break down the investment portfolio return over the past five years; equities, fixed income etc. So I just went back and took the year-by-year returns of the equity portfolio from that table going back.
And for whatever reason, that five year and ten year return happens to not match the five and ten year returns they mention in the letter to shareholders in the 2013 annual report. There must be some valid reason for that, but I don't know what that is. It's not the biggest deal, either, so...