A few days ago I wrote about my favorite 8 ideas - briefly - which was fun since people actually read it and commented on it. Unlike, say, my top short idea for 2016 which took a lot more work; work which apparently translates into more droning on and on and on than people have an appetite for. (Good news though: the stock I'm short has risen 25% in the week since I published, so you (but not me) can get a better entry point!)
Anyhow, I run a concentrated portfolio so the bar for inclusion is quite high, but if you're forced by rules or silliness to own 20-25 stocks, I think I've got lots of ammo for that.
More seriously, I've got a pool of stocks I know well - selected for simplicity above all else - that I rotate amongst. Each is, like I said, simple, and unique is some hopefully sturdy way, so that the unique thing is still there in 2017, 2018, whatever when the price gyrates into my range. Here's 3 of them, plus 1 kinda unique fixed income situation - yielding 8.25% and almost perfectly safe I think.
Parke Bank (NASDAQ:PKBK) $12.50: This bank is from Chris Christie's state but they're shockingly lean, especially for a tiny community lender in a high-cost area. The relevant numbers wiggle a lot in the short-run but right now Parke spends ~$16m annualized on overhead, and collects about $4m in fee income, so "net overhead" (think of it as a money management fee) is ~1.4% of their $885m in assets. Typical peers spend 2.5%.
[The nice things about overhead compared to something like loan losses are, a) its persistent for years, lean stays lean, and fat stays fat, whereas credit can change quickly, and b) it's not fudgeable, i.e. salary is salary whereas the process for charging provisions for credit losses is anything but.]
The gap between PKBK and peers is a huge deal. All else equal those 110 bps mean an ROA differential of 1.1% pre-tax, and 0.7% after-tax. To translate that advantage into ROE multiply by 10 (typical leverage). That's 7 percentage points! (By the way, to measure leanness I prefer net-overhead/assets to the more commonly used "efficiency ratio" - which is gross overhead divided by revenue; here is Parke's - because it's so much easier to translate into the numbers that matter to us, i.e. ROA and ROE.)
Of course that 7 point gap assumes all else is equal, where all else includes: asset yield, cost of funds, and loan loss provisions.
Parke's ROE was 11-12% the last two years, which makes all else not look equal, i.e. peers can't be doing 4-5% ROE's, can they?
Well in fact many community banks were/are so inefficient - had so little earnings horsepower - that 4-5% was their sustainable reality in the best of times; that's why the sector is consolidating. If you can't make 10% even in good times it's time to exit.
But it's true that Parke's ROE has been held back by persistently elevated credit losses, both loan charge-offs, and the expenses of preparing foreclosed real-estate for sale. Factors exacerbated by an atrocious local regulatory environment: I've heard the foreclosure process takes about a year in Texas and 3 years in New Jersey.
Luckily earnings horsepower can make up for a lot of bad stuff and Parke's got it. They should keep earning ~10-12% ROE's per year at a minimum and will be capable of a few points more than that once they finish cleaning their books, and if recession hits they can take a punch and stay profitable; indeed, they did. Trading near 1x book it's worth a spot on a watchlist.
Core Molding (NYSEMKT:CMT) $11: Core makes truck sidings. Rough economics, right? Yes but at least it's simple. The worst industries are the super competitive but sexy ones. I'm reminded of Bruce Greenwald's book on the empire building value-destroying moguls in consumer media, or my own research into fiber optics components (my aforementioned short OCLR participates), an industry that's cumulatively lost money since its inception despite creating incredible consumer value, i.e. broadband internet can't work on copper and electricity.
Shooting for the moon is more fun than trying to be a few points more efficient than the competition, so if the industry is competitive, it's best that the moon is perceived to be unobtainable.
There are a few other neat features here: truck sidings are heavy compared to their cost so truck companies like to be near their suppliers, limiting overseas competition and creating a bit of lock-in (Core has moved a plant at a customer's request before; as opposed to say, being replaced); the capital required to make sidings is expensive so hypothetical entrants think twice; and these products are all mature in their basic functionality (sure there's grinding progress in strength, weight, durability) so there's little change of the sort that obsoletes that $10m tool you bought last year.
Core's a North American heavy-duty manufacturer - mostly Mexican and a little rust-belt. Among firms that fit that description we're used to seeing atrocious records, but buying CMT at most points in the last 15 years has paid off nicely. They deliver ROE's above 10% most years, reinvest for growth, and have avoided serious busts and writedowns (profitable in 2008-2009) because they're so lean.
Three customers - Navistar (NYSE:NAV), Paccar (NASDAQ:PCAR), and Volvo (OTCPK:VLVLY) - make up most of their revenue. Isn't concentration a bad thing? In some ways, but let me point out a nuanced benefit.
It's cheap to manage few customers [same logic applies to Fabrinet (NYSE:FN), 3-4% overhead, which I've written up a few times]. SG&A is one part fixed stuff - compliance, accounting, etc. - and one part stuff that's proportional to how many relationships you've got to manage.
Now, if you forgo those benefits - if your revenue is 3 big guys, plus 15 tiny guys, plus another 30 you're chasing - you suffer the costs of concentration (risk, lumpiness) and your bloated. Bad.
But Core couldn't even screw up in that way if it wanted to - there's only a handful of trucking OEM's. Like I said, competitive but boring isn't so bad.
Core had an incredible year in 2015; earned ~$1.65, and grew fast on a strong industry tailwind and ramping of their Volvo relationship. The industry will weaken in 2016 and EPS will fall to $1-1.35 I assume. At $11 (down from upper $20's; the only way you get hurt in CMT is buying it expensive) that's more than priced.
Nicholas Financial (NASDAQ:NICK) $10.45: NICK's earning ~$1.40 on ~$12.80 book, so the valuation hurdle is set low here, especially since their current 11% ROE is way lower than what they've historically achieved through full cycles.
They're a subprime auto lender and they're really, really good at it. A quick glance at 2000-2015 suggests an average ROE of 15% with just a few instances below 10% and many near 20%. Why so good historically? Why deteriorating now?
I get the sense subprime auto lending is almost impossible to scale. To give you an image, NICK's got ~70 branches in 16 states but only ~330 people, so 4-5 per branch, who are tasked with handling dealers, borrowers, etc. in a 25-mile radius. Each branch handles only ~$2.5m in receivables.
This is a heavily regulated politically hot industry that deals in collateral that depreciates rapidly. Throw in extreme decentralization and risk/accounting/regulatory compliance looks like a massive headache that's nowhere near worth the cost. "Hey Mr. Dimon, how about we open 100 new subprime auto branches; in a few years we might do $250m in volume; and/or end up mentioned in Bernie Sanders stump-speech since we charge 25% interest rates."
So that gives a structural feel for why NICK's delivered some absurdly high and consistent ROE's for decades (using low leverage too!). People-wise, their founder and former CEO wrote President's letters in the sort of candid style Buffett's always urging too. (Reminds me of Jim Clayton, founder of the - formerly - hugely successful manufactured home company, whom Buffett raved about, i.e. lowbrow industry, sturdy culture surrounded by shady peers, very profitable, etc.)
Anyhow, NICK is pissed. Traditional lenders - the ones that got us in this mess in the first place! - are apparently moving hard into NICK's space, which is nudging all their key metrics - yields, losses, etc. - in the wrong direction. Thus NICK's terribly disappointing 11% ROE.
I imagine this will end in the usual way. The noobs will hurt themselves, and NICK too in the process, but the latter will emerge from the mess intact and prosper yet again. Or they'll get acquired by the crazy noobs.
A patient investor could buy NICK today - the price is fine. A less patient one (or one with too many other good ops - me) will wait for either an absurdly cheap price or clearer evidence of reversal of competitive pressure.
FULLL (3 L's!!), the preferred stock ("notes" technically) of Full Circle Capital (FULL), $25: Full Circle Capital is a BDC. These things (BDC's, the equity) are regulated like REIT's - not taxed but must pay out all earnings, and keep a certain fraction of assets to supposedly underserved market sectors. Apparently medium sized private American companies are underserved! Who knew?
BDC's can only take 2/1 leverage; $100m equity, then $200m in assets. Regulatory pressure might up that a bit, but not much.
It's pretty hard to make money much less blow yourself up as a lender with 2 or 3 to 1 leverage. That stinks for the BDC equity: you've got some of these companies lending at 10% financed by borrowing at 7% plus 2-2.5% management fees!!!
But it's a sweet deal for preferred stock holders.
FULL's preferred stock FULLL has an 8.25% coupon, trades at par, is redeemable this year, and matures in 2020. Full details here.
FULL is not a healthy company, but it's got one thing I like a lot: simple assets.
Many BDC's plunk up to 30% of their assets in CLO's, usually the equity of pools backed by high grade debt. In contrast FULL's got nothing but plain vanilla loans. No hidden leverage.
The only way I see a BDC hurting itself badly enough - i.e. a 50% blow to assets - to threaten the principal of its preferred stock is with complicated assets containing hidden leverage. Sure, CLO equity with 10% subordination in high grade debt is probably safe, maybe even smart, but I don't do structured finance, and I certainly don't lend to companies that do.
I'd much rather lend to one like FULL - not a good company, but a vanilla one.
FULLL is one of half a dozen preferreds/notes issued by BDC's that look nice and vanilla and thus, far as I can tell, perfectly safe, which yield 7-8.25%.
I think the inefficiency here is that BDC's are relatively new and they borrow small amounts of money. Newness, lack of statistical track record, plus issue sizes that are too small to rate much less research, are why FULL pays 8.25% rather than 5.5%, I think.
This one, and these in general, are not taxed advantaged, so they're ideal for retirement accounts.
If you're scared of the market FULLL's a retirement account option with, I think, higher expected returns than the S&P 500, with extremely low risk.
If I'm missing something with these BDC's - look too good to be true, right? - let me know!
Disclosure: I am/we are long FULLL.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: I am long FULLL, the preferred stock, and may trade long positions in CMT. No current position in PKBK and NICK.
Editor's Note: This article covers one or more stocks trading at less than $1 per share and/or with less than a $100 million market cap. Please be aware of the risks associated with these stocks.