Fairfax: The Big Short Redux?

Fairfax Financial (FFH.TO/FRFHF) is a Canadian insurer founded in 1985 by the current CEO Prem Watsa. Fairfax uses the Berkshire playbook of using low-cost float, and investment leverage to compound book value at remarkable rates.

Here is a picture I snipped from the 2016 shareholder’s meeting presentation.

Screenshot from 2016-08-27 23:00:17

From inception, FFH has trounced the S&P compounding book and market value at nearly 20% for 30 years.

It pools float derived from its insurance operations, equity, and some debt into an investment portfolio.  Currently, the investment portfolio is about $1300/share, while book value is near $400, for an investment leverage ($1300/$400) above 3.

BVPS Growth = leverage * investment return

Screenshot from 2016-08-27 16:39:12

As seen from the chart above the return on the investment portfolio has been quite volatile. Due to the volatility the compounded return (geometric mean) has been lower (~7%). FFH’s cost of float has been non-zero, around 1%. We can understand its historic returns by multiplying (investment returns – cost of capital) * leverage. Using simple numbers, this implies (7% – 1%) * 3.5 = 21% compounded growth in BVPS.

Comparison with MKL and BRK

While Fairfax seeks to emulate the business model of MKL and BRK, it is different in several respects:

  • FFH’s insurance operations have historically been lousier than MKL or BRK. Over most years the insurance operating business has lost money. Their cost of float has therefore been higher than BRK and MKL. There are some signs that this is changing in recent years.
  • Prem Watsa often takes directional macro-bets/hedges. You can see this in the chart above. His returns suffered before the dot com bust, but thrived in the aftermath. He bet big against the 2008-2009 collapse (using S&P puts and CDS), but that took a toll on his performance in 2003-2006, when everybody else was doing great. He currently has another big bet on, which has been a drag since 2010.

We can compare the BVPS growth of FFH with MKL and BRK over different time horizons.

Screenshot from 2016-08-28 11:15:27

Since 1986, all three have compounded BVPS at astounding rates. We can see, however, that BVPS growth at FFH has been quite lumpy, compared to BRK and MKL. The performance in the recent past (3-5 years) makes FFH looks quite bad.

Recent Past

Since 2010, Prem Watsa put on inflation linked hedges, fearing a Japan-style deflationary spiral. Like CDS, FFH buys derivatives that give them exposure to some notional amount (currently of the order of $80-100B).

Say the inflation level (CPI) falls by 5% before the hedges expire. Then, FFH collects 5% of the notional exposure amount ($4-$5B). The trade would be worth this profit minus the cost of all the derivatives.

As the Brooklyn Investor points out the risk-reward structure of this is a massive bet/hedge. FFH pays about 1% in ROE for the hedges. If the contracts expire worthless, then that 1% is gone. If, on the other hand, deflation does rear its ugly head then the upside could be 50-100% of BV.

Prem Watsa doesn’t need runaway deflation like the Great Depression or Japan for the bet to pay off. If it does,  FFH will end up with plenty of liquidity precisely at the time when everyone else is screwed.

The bet is very asymmetric. It begins making sense, if you think the odds of a 5% deflation over the next 5-10 years are greater 10%. Psychologically, it is extremely hard, of course. Seeing real money consistently bleed out of the door triggers a primal response. Stop the bleeding. Bring out the tourniquet.

However, let’s assume FFH is not the only stock you own. If your portfolio is anything like mine, most holdings would do poorly in a deflationary environment. FFH provides some real diversification benefits.

It hasn’t helped that FFH’s equity investments (BBRY anyone) since 2010 have turned out terrible. But some context here is useful.

Screenshot from 2016-08-27 16:40:25

The equity exposure in FFH’s investment portfolio is quiet small (22%).

Valuation

Finding the intrinsic value of FFH is harder than valuing MKL or BRK because of optionality of the macro bets. In the analysis below, I assume that the deflation bet doesn’t work out, and costs 1% in ROE until they expire.

Since 2000, FFH’s average P/B ratio has been 1.27. If we apply this multiple to the current book value of $406/share, we get a fair value of $515/share.

FFH’s management has repeatedly asserted that their target BVPS growth rate is 15%. Historically, they have managed to beat this hurdle over long periods of time, and perhaps we can trust them to do so in the future – regardless of how uncertain the present time feels. If they manage to return to this level, then FFH would probably deserve a P/B multiple of 1.5. This implies a target value of 1.5 * 406 = $609/share.

How plausible is a 15% BVPS growth rate? The key ingredients are return on investments (historically 7%), the cost of investments (historically 1%), leverage (currently 3.25), and the cost of the equity hedges (~1%).

ROE = (investment return – cost of investment) * leverage – cost of hedges

Currently investment returns have been around 2%, the cost of investment about 1%, a leverage of 3.25, and cost of hedges around 1-1.5%.

Screenshot from 2016-08-27 16:41:20

This has implied an anemic ~2-3% compounding in BVPS in the last few years.

However, if we believe that the present is an anomaly, and normalized performance will be superior, we can use our best guesses for the numbers.

To get 15% ROE, with a 1% cost for hedges, and leverage of 3.25, FFH needs to ensure that (investment return – cost of investment) = (15-1)/3.25 = 4.3%.

How difficult is this?

In recent years, FFHs insurance operations have improved significantly. In 2015, for example,  FFH had an underwriting profit of $705m. The combined ratios at most insurance subsidiaries have fallen. Currently $17B out of the $29B investment portfolio is financed by float (~60%; the rest is financed by equity and debt). If insurance operations turn consistently profitable, then the cost of investment could well drop to or below zero.

Over long periods of time, FFH has been able to compound its investment portfolio at about 7%. So 4.3% looks very doable.

Conclusion

FFH’s fair value seems to be between $515-$610/share. The hedges provide optionality and diversification. The narrative here is that of an insurer with improving operations, temporarily depressed investment returns, with a huge asymmetric bet on deflation, run by a foxy CEO with skin in the game and history of successful (albeit early) macro bets.

Useful Links

  1. FFH 2o16 Annual Meeting Presentation
  2. WertArt Capital blog on FFH
  3. Ben Hacker at Remick Capital 2009 commentary
  4. Leisure Capital’s reality check on BVPS growth
  5. Defy the Crowd Podcast Episode on FFH
Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s