Markel at Book?

This blog is getting boring, I know.  All these financial companies with great historic track records at book value (JPM, GS, BRK).  An incredible hedge fund manager available to the public (GLRE).  Yawn.  Nobody seems to care.

But good investing isn't supposed to be exciting.  Sometimes, it's boring.  There's nothing to get excited about.  The outlook stinks.  Financials are risky.   Yes, yes, yes.  I wouldn't advocate anyone putting too much of their equity exposure into financials.  Even the folks who advocate focused or concentrated investing recommends some industry diversification.

Anyway, I noticed that Harleysville Group (HGIC), an insurance company has been acquired for 2x book value per share.  We all know insurance companies (including BRK) are trading very cheap now; most of them are trading well below book value per share (BPS).

But this HGIC deal reminded me of a comment that Steve Markel of Markel Corporation made a few years ago, granted this was before the big blowup in subprime and "Great Recession".  He said that Markel Corporation is worth 2x book value.

Well, MKL closed Friday at $357 per share versus a BPS at the end of the second quarter (June 2011) of $338.66 or 1.05x book value.  OK, so I cheated.  It's not fair to compare the stock price now versus the book value of MKL as of the end of June 2011; the stock market has declined substantially since then and MKL owns a bunch of stocks!

You got me.  So let's adjust it.  MKL owned $1.84 billion in equities at the end of the last quarter and stock prices have declined -14.3% (as measured by the S&P 500 index).  Let's forget that MKL's equity portfolio has outperformed the S&P 500 index in just about all time periods (we'll look at that later) and assume that the MKL equity portfolio went down as much as the S&P 500.  That would be a loss of around $260 million.  Let's keep it simple and forget about possible underwriting profits since then and interest income on their $5.5 billion fixed income portfolio.

A $260 million loss would translate, at 9.76 million shares outstanding, to around $27/share.  So knock that off the most recent published BPS of $338.66 and you get around $312/share in BPS.   Versus the current $357/share stock price, the BPS is 1.14x.

Not that bad at all.  It's a slight premium.

Is MKL worth 2x BPS?
Well, let's see.  Below is a table of MKL's BPS and closing price (at year end) since 1990.

            BPS          Price           P/B ratio
1990    $10.27       $11.75         1.14
1991    $15.59       $22.00         1.41
1992    $20.24       $31.25         1.54
1993    $27.83       $39.38         1.42
1994    $25.71       $41.50         1.61
1995    $39.37       $75.50         1.92
1996    $49.16       $85.00         1.73
1997    $65.18     $156.13         2.40
1998    $77.02     $181.00         2.35
1999    $68.59     $155.00         2.26
2000   $102.63    $181.00         1.76
2001   $110.50    $179.65         1.63
2002   $117.89    $205.50         1.74
2003   $140.38    $253.51         1.81
2004   $168.22    $364.00         2.16
2005   $174.04    $317.05         1.82
2006   $229.78    $480.10         2.09
2007   $265.26    $491.10         1.85
2008   $222.20    $299.00         1.35
2009   $282.55    $340.00         1.20
2010   $326.36    $378.13         1.16

Five-year average P/B:  1.5x
Ten-year average P/B:   1.7x
Average since 1990:      1.7x

So judging from past experience, it looks like 1.1x book is pretty cheap for MKL, as cheap as it's been since 1990.   I'm tempted to say look what happened to the stock the last time it traded at 1.1x book (1990), but I shouldn't.  Things may be different now so we can't just expect the same sort of growth and stock price appreciation MKL saw from 1990 to 2000.

But we can say that MKL is pretty darn cheap at 1.1x book given both MKL's successful track record and historical valuation.

Of course, some will argue that the trailing five year growth in book value per share in the late 1990s was well over 20%/year versus a more subdued +14% in the most recent five years.  That's why, they'll say, MKL was worth more back then than now. Well, this sort of makes sense on the surface, but doesn't really.

The very high returns posted in the late 1990s was partly due to a really strong stock market that took stock market valuations to very high levels.  So to put a high P/B multiple based on historically high performance may not have been as rational as it may have seemed at the time.  Why?  Because the higher the stock market valuation and higher the returns posted by MKL at the time, the higher the probability that the stock market was overvalued and prospective returns would be lower.

So effectively, what the market did to MKL stock back in the late 1990s was to put a high multiple on LOWER prospective returns.  This is the problem with investing with a rearview mirror.

Today, historical stock performance looks pretty dreadful, at least in the last five to ten years.  And MKL's book value growth has been good, but not as great as it was during boom times.  So the market values MKL much lower. 

But we may be in an opposite position as 1999/2000.  Maybe the prospective returns in the market now is HIGHER, but the market is valuing MKL much LOWER.  Again, because the market is investing with a rearview mirror.  They are looking at historical returns, not prospective returns.

With that view, perhaps MKL is much more attractive than it seems.  Admittedly, P/B does also correlate to investment leverage; the higher the leverage, the bigger boost to BPS investment returns have.  But with an equity portfolio of over 50% of shareholders equity currently, a 10% return on that portfolio will boost BPS by 5% or 3.25 after tax, and a 20% return would boost it 10% pretax and 6.5% after tax.  It is fair to argue that investment leverage has a higher contribution to BPS growth than equity returns.

Another issue is that the P/B ratio history looks a lot like Berkshire Hathaway's (and many other insurance companies).  However, MKL does not have a "Buffett is getting old" problem.  Some argue that the P/B of BRK is declining due to the age of Buffett (the closer we get to the end of his tenure, the lower the valuation; maybe we can calculate a mortality table adjusted P/B ratio for BRK).

MKL also doesn't have the size problem of BRK.  BRK, at more than $150 billion is pretty big.  MKL only has $3.2 billion in shareholders' equity.  Puny.

Also, most insurance companies invest only in fixed income securities and with ever decreasing interest rates, the value of the 'float' continues to decline.  This expected return on equity for many insurance companies can get very low due to lower interest rates.

However, MKL differs in that regard as they do invest much more in equities.

So are they good equity investors?
OK, so that leads to the next question.  Are they any good picking stocks?  Should an insurance company take equity market risk?

Markel's equity investments is run by a value investor named Thomas Gaynor.  You can google him and read a lot about him.  He has done really well. He sticks to high quality American blue chip stocks and concentrates his investments (as opposed to buying 1000 stocks like a closet indexer).

Here are the equity portfolio annualized returns for the five and ten year periods versus the S&P 500:

                               MKL's equity portfolio         S&P 500
Five year                 +6.2%                                  +2.3%
Ten year                  +7.6%                                  +1.4%

These returns don't look like much to be sure.  But beating the index by 3.9%/year in the past five years and 6.2%/year over the past ten years is quite a feat.  Go ahead.  Do some quick searches on Morningstar and see how many mutual funds you can find that did that.  You won't find many.  If you do, maybe you should invest in that fund.

But wait.  The equity portfolio return isn't the only that matters here.  An insurance company gets premiums from clients to invest.  This is called float.  If you can make decent returns on that float, that boosts your return on equity.

Notice how MKL's equity portolio only returned 6.2%/year in five years and 7.6%/year in ten years.  How much did MKL's BPS grow in the last five and ten years?

                                Change in BPS of MKL
Five year                   +13%/year  
Ten year                    +12%/year

That's pretty good.  Look at BPS growth versus the S&P 500 index; much higher. 

Part of it was due to the performance in the stock portfolio with some input from the fixed income portfolio and what is called float leverage.

Let's see how the whole investment portfolio of MKL has done over the past five and ten years:

                                               Five years                 Ten years
Equity portfolio                        6.2%                         7.6%
Fixed income portfolio             5.3%                         5.5%
Total portfolio                          5.4%                         6.0%

So MKL boosted BPS by 12%-13% with portfolio returns of 5.4%-6.0%.  Again, that's float leverage.

Isn't leverage risky?  Yes, it can be.  But MKL invests most of the 'float' amount, in other words the amount owed to policy-holders in lower risk investments and only invests some portion of shareholders' equity in stocks.  This is getting too long so that will have to be another post some day.

But are they any good at writing insurance?
Before we finish up, this is an insurance company.  We have ask ourselves, hey, is this company any good at writing insurance?   Fair question.  In the insurance world, the standard measure is what is called the combined ratio.  If it is over 100, they are losing money in the insurance business and if it is under 100, then they are making a profit in the insurance business before investment income/returns.

Let's see how they measure up historically.  This is the combined ratio of MKL compared to the industry.

                  MKL           Industry average
2006          87%               92%
2007          88%               96%
2008          99%             105%
2009          95%             101%
2010          97%             103%

So, that looks pretty good to me.  I think their history for longer time periods were pretty good too.  You can get all that info in their annual reports (which you should read before considering buying this stock, of course, instead of relying on some anonymous guy on the internet.  Never buy a stock you read about on the internet without doing some of your own homework!!)

Some will argue, fairly, that much of the underwriting/insurance profits in the past few years by MKL and other insurance companies have been due to favorable reserve development.  That means they are releasing reserves (amount put aside for possible losses on insurance policies) into profits which makes their insurance business look better than it actually is.

Well, this can be seen in two ways.  Yes, if that is the only source of profits and the insurance market doesn't improve, profits and combined ratios going forward can get worse to be sure.  But if we assume they are conservatively writing business and pricing it correctly, it shouldn't be too much of a concern.

(By the way, the insurance market has been in a 'soft' market for a long time now, but that's another long post for another day; besides, I am not an expert in this industry).

An interesting counter argument to the above is this: What would people be saying if they had to keep adding to reserves to previously written business?  Bingo!  They would argue that they have been reckless in writing policies and underreserving them and therefore overstating profits for the past few years!   Pretty much what happened to AIG. 

So critics will complain if you have to build reserve (and reduce profit) or release reserves (and increase profit).

In any case, the insurance business is like the banking business.  We can stare at the GAAP (generally accepted accounting principles) figures all day long, but at the end of the day, it all boils down to management and the people.  You have to trust the people running the business.  If you don't, it doesn't matter what GAAP says.  If you do, you don't worry too much. 

MKL does tend to have a reputable history.  Here are the average combined ratios averages over time:

                                           MKL combined ratio
Five year average               93.2%
Ten year average                99.0%
Average since 1989            98%

What about catastrophe risk??
Good question.  This is an insurance company.  When outlier events happen, all hell can break loose.  But the way I see it, they have managed to increase book value for the first six months of this year through June-end 2011 despite having to book losses for the U.S. storms, Japanese earthquake and tsunami, Australian floods and the New Zealand earthquake. 

That's a lot of damage.  To me, that's sort of a nice stress test.  And they have grown book value despite those losses.

Also, look at the history of BPS since 1990 above.  That includes all sorts of stuff, like 9/11, Katrina and other huge disasters.  I think they have done well through those big events.

Financial market risk?
Again, if you look at the BPS since 1990, you will see that they have gone through a LOT of real-life stress testing and passed.  I won't list up all the things that happened, but a bubble popped in 2000, we had a pretty serious near-depression/financial collapse, and they have come through it pretty well, I think.

Unless you think we are at the beginning of a real depression, MKL seems pretty solid to me.

What is the normalized BPS growth for the company?
OK, so the last thing I'll look at is, what is the normalized ROE or BPS growth of this company?  We know we can buy it at 1.1x book.  But what is MKL going to earn on that book?  That is the key, isn't it?

Yes.  So let's look at that.

Let's first assume that MKL can, over time, at least break even on their insurance business.  Given the long term averages I posted above, that seems reasonable.  And then let's take really conservative return figures for their fixed income portfolio and equity portfolios.

Let's say the fixed income portfolio can earn a 4% yield and the equity portfolio can earn 5%.  That seems pretty conservative.  Using the portfolio figures from June 2011, that would be a net profit of:

Fixed income portfolio of $5.5 billion x 4% interest = $220 million
Equity portfolio of $1.84 billion x 5% return = $92 million
less interest expense on debt of $74 million = $238 million

For a pretax return on equity of 7.3% or a an after tax return equivalent of 4.7%.

This doesn't look too exciting.  Clearly, if equity returns aren't attractive going forward and interest rates don't go up or it keeps going down, then MKL may not be such an interesting investment.

However, there is one more factor that determines the return on equity of MKL.    That is investment leverage.  The good news is that in a soft market, MKL is not writing business left and right at bad prices (that would lead to large losses later on).  Gross written premiums were 0.6x shareholders' equity in 2010 versus a five year average of 0.86x and a ten year average of 1.3x.  Since 1989, this was 1.9x.   So we know they are being conservative in writing business; they write less business when prices are low.

This leads to lower investment leverage.  Investment leverage now as of June 2011 was 2.4x versus a five year average of 3x, ten year average of 3.3x and 3.8x since 1989.

To summarize that,

                                   Premiums written to total equity             Investment leverage
Current                        0.6x  (2010 full year)                               2.4x (as of June 2011)
Five-year average       0.9x                                                           3.0x
Ten-year average        1.3x                                                           3.3x
average since 1989     1.9x                                                           3.8x


In order for MKL to achieve higher BPS growth, interest rates will have to go higher, equity returns have to go higher and investment leverage must increase or any combination of the above.

The growth in BPS in the past five years has been good despite low investment returns due to the 3.8x investment leverage.

Total investment return for the five and ten year periods through 2010 were 5.4% and 6.0% respectively.

At the current 2.4x investment leverage and a 5% investment return, BPS growth would be 6.4% (net, after tax).

However if investment leverage goes back up to 3.0x, then at a 5% investment return, BPS growth would be 8.3% and at 3.5x it would be 9.9%.

Here is a table that summarizes the future book value growth assuming different investment returns and investment leverage levels:
      
                     Investment return:
leverage       4%          5%            6%
2.0x             3.73%     5.03%       6.33%
2.5x             5.03%     6.66%       8.28%
3.0x             6.33%     8.28%     10.23%
3.5x             7.63%     9.91%     12.18%
4.0x             8.93%     11.53%   14.13%

The above table shows the total after tax change in book value one can expect assuming a combined ratio of 100% (break-even insurance business), various leverage and return levels.

For example, if leverage due to a stronger insurance market gets back up to 3x and total pretax investment return is 5%, then the BPS of MKL should grow 8.28%.   This sounds very reasonable.

However, if a soft insurance market continues and equity and fixed income returns heads down, then assuming a 2.0x leverage ratio and a 4% pretax return on investments would lead to only a 3.73% increase in BPS.

It's important to keep in mind that this is a static analysis and does not include the possibility of acquisitions or share repurchases by MKL.    This is an important part of growing the business and/or BPS.  The above analysis also doesn't include the the Markel Ventures business, which is an operating business apart from the insurance business.   Like at Berkshire Hathaway, this non-insurance operating business may grow substantially, but that is not reflected in the above figures so I would consider the above very conservative.


Conclusion
In any case, for BPS to grow in double digits, it does look like MKL will need to increase investment leverage and/or returns.  Continuing investment returns at a 5-6% pace of the past five to ten years with lower investment leverage will lead to lower growth in BPS (at the current 2.4x leverage, BPS growth should be in the 6%-8% range with 5-6% investment returns).

I admit this static analysis is not as exciting as I thought it would be;  I was hoping that I could say even given current returns and leverage, BPS would grow double digits.  But that's not the case.  The market does seem to be pricing in a soft insurance market and poor investment returns for the foreseeable future.

Any change to that could see some upside in MKL stock.  Since the P/B ratio is correlated to investment leverage and returns, any bounce in the insurance business (harder market) and increasing float/leverage along with improving returns on investment can really leverage the upside in this stock (BPS growth due to better returns with a higher multiple on that growing BPS on the stock for double whammy potential).  The downside would seem limited at close to book value.  

Call it a call option on an improving insurance market and the continuing success of MKL's management in growing book value over time.