In my previous post I valued Berkshire Hathaway using Tom Gayner’s valuation method, which he described in a Value Investor Insight interview from 2012. So I figured I might as well put Markel’s numbers into the spreadsheet and see what we get. Before I get to that here’s a quick summary on Gayner’s valuation method for those who didn’t read the last post.
Valuation Method Summary
Gayner breaks down the business valuation into three parts:
- underwriting profits
- investment returns
- non-insurance normalized earnings
He then applies a range of returns to insurance premiums earned, the investment portfolio and also estimates normalized earnings for the operating businesses. Finally he takes the sum of all three parts and applies a multiple of 10x, 14x and 18x. I changed things up a bit, but most of it is self explanatory.
Markel’s Investment Returns
I decided to adjust the investment portfolio returns down (vs what I used in the Berkshire model) to a base case of 5% with a low of 3% and high of 7%. Historically Markel has achieved higher investment returns (see below):
- 5-yr annual return: 7.4%
- 10-yr annual return: 6.2%
I’m being conservative with Markel for a few reasons.
1. Markel has an unusually large fixed income allocation due to the acquisition of Alterra
When the investment portfolio from Alterra was inherited by Markel the total investment portfolio went from $9.3B in 2012 to $17.6B in 2013. It almost doubled! Alterra’s $7.9B investment portfolio consisted mostly of fixed income securities and a hedge fund portfolio that was subsequently liquidated. This diluted Markel’s equity allocation, which currently sits at roughly 55% of shareholder equity. Obviously having less invested in equities and more in bonds will lead to lower returns. Historically Markel’s equity returns have outperformed fixed income returns by a wide margin.
- 5-yr annual return: Equities, 20.4% Fixed Income: 4.7%
- 10-yr annual return: Equities, 13.4% Fixed Income: 4.6%
2. Markel doesn’t use float to invest in equities like Berkshire
Markel invests in equities with shareholder equity only; whereas, Berkshire uses both shareholder equity AND float. Markel’s float is mostly invested in fixed income securities that match Markel’s insurance liability durations. In a Gurufocus interview from 2011 Gayner explained why Markel does this vs. investing in float like Berkshire:
TG: “Berkshire has a much bigger balance sheet than Markel and a much bigger base of equity compared to its insurance liabilities than we do. As such, they can allocate more of the investment portfolio towards equities. Over time, if we continue to grow and build up our equity capital we should be able to move in the same direction as Berkshire. Furthermore with a rising interest rate headwind and potentially lower forward returns from stocks I’m lowering expectations for conservatisms sake.”
I thought that last part was very interesting. I’ve seen Gayner state that his preference is for equities to be 80% of shareholder equity, but I’ve never heard him say they would like to eventually invest the float. This strikes me as another under appreciated benefit to the upside that could juice returns further as Markel grows in scale. For now though, returns have and will most likely be lower than Berkshire’s.
3. Rising interest rates
With such a large fixed income portfolio the obvious drawback is as interest rates rise the existing fixed income values will fall. And theoretically equity returns should be lower in the future. This combination would put a drag on investment returns going forward. On the other hand as rates rise (assuming it continues) Markel’s investment income will rise too. But for the sake of conservatism I’m lowering return expectations.
Markel is being proactive by keeping durations lower than normal and waiting.
From the 2014 Annual Letter:
“We continue to own a portfolio of fixed income securities which mature faster than what we expect from incoming insurance claims. We will continue to maintain this modest override from our normal design until such time as interest rates are higher than current levels. We just don’t think we are being paid appropriately to take the risks of owning long-term bonds so we won’t do it.”
On rates rising over the long term:
“We normally don’t try to predict interest rates but we can use common sense to say that we believed they were too low during the last few years, and now they are trending back to a more normal level. Consequently, we too will trend back towards a more normal bond portfolio over time. This should increase our investment income substantially in the years to come.”
Again, a little off subject, but Gayner and co. are aware of the rate environment and as with insurance underwriting they are being disciplined by waiting for a price that warrants the risk taken.
Markel’s historical combined ratio is 96%, they had a 95% combined ratio in 2014 and so far in 2015 (through Q3) have an 89% combined ratio. Due to fairly consistent underwriting profits I kept underwriting returns at 4%, the same as Berkshire’s. I used 2014 net cash earnings for Markel Ventures opposed to Gayner’s use of EBITDA; and as previously discussed, I lowered investment returns to 5% for the base case.
The base case shows a FV of $1,215 per share with 38% upside from Tuesdays close of $879.85. Fair value here would imply 2.2x book value, which is at the upper end of Markel’s historical range. Seems a little rich, but the assumptions aren’t overly ambitious. If you play around with the assumptions you’ll see (especially when compared to Berkshire) just how much investment return drives the overall intrinsic value for Markel.
At base case $81 EPS Markel trades at:
- 10.9x comprehensive earnings
- 14.7% ROE
If we take the base case assumption down to 4% investment returns we get a FV of $1,014 (15.3% upside). If we take it down to 3% we get a FV of $814 (-7.5% downside). Of course you can also play with the multiples. If Markel is only getting 3% returns on investments does it deserve a 15x multiple? So for me this one is a bit tougher than Berkshire. The range of outcomes are greater no doubt.
A few other thoughts on Markel’s bright future:
1. Markel Ventures growth
I didn’t forecast any growth for Ventures in the model because a) disclosure is minimal and b) it’s still a fairly small part of Markel. EBITDA for 2014 was $95m which comes out to just 8% of the estimated total net profit. Also, it’s just too hard to forecast (at least for me). But we can still look at what has happened so far. Ventures has grown significantly over the years with 30% annual revenue growth since 2005. Adjusted Ebitda has grown significantly as well.
(BTW if you’re questioning Gayner’s use of EBITDA I recommend you check out The Brooklyn Investor post I linked to at the bottom)
Again, this is off a pretty small base and at the moment is still a small part of the business, but I think Ventures could play a large part in Markel’s growth over the next decade. Gayner said at a recent Markel brunch that if Ventures was outside of Markel they would earn greater than 20% ROIC. Ventures gives Gayner yet another avenue he can allocate capital to and should provide consistent cash flows.
2. The Leverage Ratio
The leverage ratio is currently 2.45x, which is at Markel’s low end historically. If leverage increases to the historical average of 3.5x (or up at all), returns will be magnified.
3. Increased portfolio mix towards equities
This one ties into the previous bullet. Gayner has said in the past that their target goal for equities is 80% of shareholder equity (currently at around 55%). This could be viewed as a nice tailwind for future returns as the investment portfolio mix moves toward equities, but that also depends on your view of the market and Gayner’s ability to beat it. Additionally, as I quoted earlier, Gayner believes at some point Markel will be able to invest the float in equities a la Berkshire. Currently Berkshire has about 65% of it’s investment portfolio in equities (rough estimate); whereas Markel has only 23%. Bottom line, an increase in equity allocation could be a significant tail wind for returns in the future.
I do think that Markel’s stock could take a breather and pull back some. Berkshire did just that in 2015. It peaked around $150 after a pretty impressive multi-year run up and then receded back down to the $127 range (now ~$131). If Markel is a large percentage of your portfolio than maybe trimming wouldn’t be a bad idea. I did this with Berkshire when it was around $150. I sold 20% of my shares as it was encroaching on 13% of my portfolio, then Berkshire stock dropped down to $127-128 where I bought the shares back. A rare and lucky “trade” for myself. This approach allows you to take a little off the table in the short term, but still participate in the long term compounding of a great business. That said sitting on your hands might be the best bet.
In my original post on Markel I had a FV range of $800 to $1200 using a few different valuation methods. I’m pretty comfortable saying that at today’s price Markel is at least fairly valued. I know that opinion is not very popular lately, especially on twitter. Some people are worried that Markel is getting expensive after the rather quick run up in the last year or two (65% return in 2 yrs). This strikes me as shortsighted if you believe Markel is truly a great company that can compound overtime. How many times in Berkshire’s history did people get burned by selling when they thought the stock was slightly overvalued. Of course Markel is not Berkshire, but I tend to believe the business model is one built to compound overtime in the same vein as Berkshire. In my opinion there is quite a bit of optionality in Markel’s future.
So that’s my two cents. I’ll leave you with some interesting quotes from Chuck Akre’s 2011 VII interview:
On rising interest rates
“Just as TD Ameritrade is a coiled spring levered to rising interest rates, so too is Markel when there’s an upturn in the insurance pricing cycle.”
On the Leverage Ratio:
“As part of that model, they constructed a holding company balance sheet which typically had $4 of investments for every $1 of book value, so that when they earned 5% after-tax on their investments, that was magnified four times and the change in book value was 20%. When pricing is soft and the business isn’t growing, that gearing ratio shrinks because they can’t add enough to the investment portfolio to maintain the 4:1 ratio. Today the gearing is more like 2.5:1.”
“So you’re paying around 125% of book. That is hardly expensive for a high-quality, well-run and transparent insurance company that I believe can compound book value at 15% annually. As I mentioned earlier, 200% to 300% of book is not at all out of line for this type of company in a different part of the cycle.”