Leucadia National (LUK), often lumped into the category of mini-Berkshires, was a Wall Street darling not very long ago. Exploits of its long-time architects, Ian Cumming and Joe Steinberg, were the stuff of legend.
They perfected the “buy – fix – sell” mantra, and applied it to businesses Warren Buffett wouldn’t look at twice (commodities, near-bankruptcies, etc.), with remarkable skill.
Then around 2010-2011 two things happened: (i) the old guard passed on the reins to a new dynamic duo Richard Handler and Brian Friedman (who I like) via the transformative acquisition of the investment bank, Jefferies, (ii) they bequeathed a motley bunch of companies that were a bet on big-time inflation, and an uptrend in commodities and China, to their successors.
We know how that bet turned out.
Results have been brutal. Everything that could go wrong has. And then some.
Interest rates decimated earnings at JEF. National Beef got crushed in a downward cyclical spiral. Oil got destroyed. China caught a cold.
The share price promptly obliged by dropping to sub-$15 in 2016 from highs near $40 in 2011, despite reasonable moves by the management.
But here is the good news. The number of ways in which the future can now surprise to the upside is now much greater than the number of ways things can head further south.
The biggest risk, in my opinion, is more muddling around.
Assets may be under-earning. But if you have faith in any form of mean reversion, the optionality in LUK packs a serious punch.
I snipped the following graphic from LUK’s June 2016 investor presentation.
As you can see, LUK contains a bunch of different business. Jefferies, accounts for nearly 40% of the book value and towers above the rest. The others can be classified into two buckets: the financial services business (mostly forays by new management), and the traditional merchant banking business (mostly legacy Cumming/Steinberg investments).
The investor presentation presents a nice overview, and the dials that management is trying to turn to drive value.
LUK’s assets are currently not pulling their weight. Even otherwise, earnings at LUK have been lumpy. So using an earnings or cash-flow based approach to value LUK is unreliable.
It is best to use an asset-based approach to value LUK. Let’s do a quick back of the envelope calculation, first. That will give us some sense for numbers.
We can follow that up with a more careful sum-of-the parts (SOTP) calculation, where we value some parts on asset value, and others on earnings.
LUK’s tangible book value (TBV) is approximately ~ $7500m. The number of shares out is 372.4m. So the TBV/share = $20. The book value (BV), which includes goodwill and intangibles, is about about 35% higher (~$27/share).
Historically, LUK traded between 1-1.5BV. Currently, it trades at TBV, which represents a substantial discount to book value. If things go back to normal, we should expect – purely on a “reversion to the mean” argument, for LUK to trade between $27 – $40/share.
That is a 40-100% upside. The downside is probably capped by the TBV (which of course could decrease).
Here is a picture of my Excel sheet, which summarizes my SOTP analysis. I’ll annotate some of the choices in the comments below.
For many parts, I use either the book value or some reasonable discount to the book value. For other segments, I look at not what they are currently worth, but what they might be, if they begin firing on a few cylinders. We can do this by looking at how comparable firms are being currently valued.
One complicating factor is that LUK often doesn’t own 100% of its holdings. The table below lists the fractional ownership in the “Own” column.
- Management has indicated that they want Jefferies to return to double digit ROE. If they manage to do so, then JEF would probably deserve to be valued at least at book. Prof. Aswath Damodaran maintains a helpful repository of P/B data for different industries. For “Brokerage and Investment Banking”, the average P/B is 1.16. So using a P/B of 1 is a reasonably conservative choice for Jefferies, which historically, has commanded a better than average multiple. JEF is where bulk of the value in LUK is.
- For National Beef, I copied the method used in this Value Investor Club writeup. Many pure-play meat-processing firms are private, so it is not easy to find good comparables. For JBS, a large meat processor, the EV/Revenue ratio is about 0.4. Tyson foods has a EV/Revenue ratio ~ 1, and Hormel Foods has EV/Sales ~ 2. The food processing industry as a whole has a P/S of 1.8. We use the conservative (and perhaps most appropriate) 0.4 ratio and tone it down further to 0.3 to arrive at a conservative estimate for National Beef, once it catches some tailwinds.
- Berkadia and Garcadia are great businesses. They sport great ROEs and growth rates. Unfortunately, they are too small to move the needle significantly. To value these parts, I take pre-tax income and apply a reasonable multiple.
- The asset management business (LAM) is valued at book. The industry average for “Investment and Asset Manager” firms in Damodaran’s spreadsheet is 1.11. So this is a reasonable choice that allows for LAM being moderately sub-par.
- In the June 2016 LUK presentation, a SOTP analysis of HRG is presented. I use those numbers instead of market value to estimate the value of HRG.
- FXCM, HomeFed, and Linkem, I value either at book or at market. It doesn’t really matter how you look at them, so long as you understand that there is quite a bit of optionality built into these valuations.
- Vitessee+Juneau is valued at 75% of book, given oil’s current woes. But if oil rises, this might be a coiled spring that can provide significant value to LUK.
- The other businesses don’t really contribute meaningfully to LUK. I am a little whimsical with the numbers I assign to them. Collectively, they contribute less than $1 on a per share basis.
- LUK has a big deferred tax asset, which allows them to protect future income from taxes. Since the asset will be monetized over time, I assume a large haircut (75%).
Based on all of these numbers, I arrive at a conservative value of $30/share for LUK.
I benefitted enormously from two previous SOTP valuations of LUK. One of them was by Ben Hacker (Remick Capital), and the other was the aforementioned VIC writeup. Both these valuations were done around Dec 2015.
For comparison, I looked at the “Base Case” valuation by Ben Hacker, and averaged the “low” and “high” valuations in the VIC writeup. The former comes up with a $32/share price tag, while the latter get $29/share, even though the value assigned to the constituent units is different.
Regardless of how you look at LUK, it seems that its intrinsic value is in the $27-$32 range – a slight premium to the current BV. Currently shares are trading at a nearly 40-50% discount to this fair value. If some or many of the factors that are holding the earnings start being resolved, LUK might prove to be a very profitable investment.