From reader after last night’s post.

I hope you are well. I think your blog is fantastic, thanks so much for sharing the time and wisdom for so little 🙂

I was wondering whether you could elaborate a bit more on the bad business models existent in the insurance field. If there would be a simple rule of thumb or similar it would be useful, but I’m guessing it has to be something (difficult ?to analyze) like chasing growth when premiums are insufficient, hiding leverage through subsidiaries, etc.

This was your comment:

Now, let me list for you the companies I would avoid on this list: IFT, GLRE, AGO, AEL, CNO, AIG, XL, MBI, LNC, FBL, AHL, ING, AXAHY, AFG, GNW. ?That does not mean that I endorse the others. ?In general, those that I say to avoid have poor underwriting skills or a bad business model.

(AIG is the biggest position in my portfolio.)

And secondly, I was wondering whether the fact that some are based in Bermuda gives them (or the LT investor) a competitive advantage when it comes to compounding (t)BV over time, because they are paying a lower tax-rate, aren’t they?

[normally, investors have to suffer a double whammy for taxes: the companies they invest in are taxed when they have profits –the US has one of the highest tax rates internationally–; and then they are taxed when realizing the capital gains / receiving dividends. Which led me to think that if one would be investing in Bermuda-based cos. through a ROTH IRA account, he would be avoiding both fronts, right?

Thanks so much for your time David.

I know it was a long email, and I apologize for that, couldn’t make it shorter…

Okay, let me take it piece-by-piece. ?I have biases, which I think are well-informed, but they are biases, ways of foreshortening the deluge of data, so that I can avoid making big errors.

1) I don’t believe that financial and mortgage insurers have an actuarially valid business model, and the last crisis proved me right. ?Thus I am not interested in AGO and MBI.

2) I don’t believe that long-term care is insurable, and so I am not interested in CNO and GNW.

3) With AIG, I don’t think that all of the reserve strengthenings are done for them. ?They have always been aggressive in reserving. ?I am not sure that has changed.

4) I think the business that IFT is in is unethical, and difficult if legal.

5) I think the takeover of AHL will fail, and the stock price will fall.

6) I think that many common life insurance reserving practices are liberal, and so I don’t like AEL, LNC, FBL, ING, and AXAHY.

7) With XL, GLRE and AFG, I don’t respect the management. ?Maybe with a few more years, that might change.

This explains my views on these companies. ?Other questions, let me know.

I try to run an ethical blog here, so when I make mistakes, I admit them. ?In this case, I don’t think the errors make a lot of difference to the investment decision, but I will confess to being wrong on ?details in my last post. ?I made the statement:

Though there are no financing contingencies to this deal, ACP Re can walk away with no penalty if it merely wants to do so.

That’s wrong. ?ACP Re can walk away of its own accord if there is a material adverse change, and under some conditions, they would receive a breakup fee. ?As such, it is not a “free look.” ?But it is one-sided in this sense: if the reserves are too low, ACP Re can declare a material adverse change. ?If they are fair or high, ACP will happily do the merger and enjoy the profits.

On the delay of the 10-K, which is more than a month late, I repeat that most of the figures in the balance sheet are easy to calculate. ?I was trained as an actuary, albeit a life actuary, though I was an insurance buy-side analyst for 4.5 years. ?The difficult question with any P&C insurer is whether the reserves are correct, and even actuaries inside a company are never fully sure of the reserves. ?That’s why reserves at P&C insurers are usually set conservatively, even though GAAP says to use best estimate. ?It is not a bad thing to bend GAAP accounting to be conservative, and be slow in recognizing income.

My experience with insurers that are tardy with their financials is that it is wise to steer clear. ?Aggressive insurance management teams tend to go through a string of corrections before the financials are set right.

Between 1998-2000, I used to do arbitrage on small deals. ?On net, I did fair with it, but the deals where I lost, you could feel a kind of “sag” where you would not ordinarily expect it. ?Good arb deals show strength after an initial period of selling by those that do not want to hang around for the arb.

Now, I don’t think my reasoning is depressing the stock price, but it is interesting that the stock price keeps heading lower, and slowly. ?I have a saying that slow moves tend to persist, while fast moves tend to mean-revert.

I don’t have any inside information, but this situation feels bad. ?Ordinarily with takeovers, the bid for stock is far more firm.

Full disclosure: No positions in any of the companies mentioned

A while ago, I wrote a piece on Tower Group after its stock price imploded, before it went down more, and attracted an acquisition offer from entities affiliated with the main owner of AmTrust Financial Services for $3/share. ?Here’s another letter, from a different respective reader:

Hello, David:

I’ve been a longtime reader of your columns (back to RealMoney) and have a lot of respect for your opinion as an investor and analyst, particularly your insights into insurance companies.

Merger arb has been a (small) part of my toolkit for the last 15 years but haven’t yet seen an insurance merger quite as complicated as TWGP’s acquisition by ACP Re and AFSI.? With an 8% discount to the offer price and about 4 weeks until the shareholder meeting, this one looks intriguing.

There’s a (wordy) analysis of the deal terms here:? http://seekingalpha.com/article/2106193-towers-merger-offers-opportunity-for-double-digit-annualized-returns .

A couple of specific questions about the deal, if you feel inclined to respond:

1.? Does Karfunkel’s potential conflict of interest in selling the part of TWGP he doesn’t want (commercial, personal) to the publicly traded company he chairs (AFSI) raise enough of a red flag that regulators may intervene?

2.? Are the NOLs owned by TWGP usable by ACP if there is a reverse takeover (TWGP the surviving entity) under Bermuda law?

3.? Does the price at which Karfunkel is selling the pieces to public entities using mostly public shareholders’ money raise any red flags to you?

As I said, I respect and enjoy your work and hopefully you enjoy it enough to continue.

First I will handle the questions. ?Then I will hand out a few opinions.

On question 1, the answer is not likely. ? The regulators will disallow any situation where an acquisition would significantly impair the ratio of capital to required capital to bear risk [RBC]. ?Now, shareholders could be another matter. ?In this acquisition, two public companies that the Karfunkel families control are buying up the renewal rights to Tower Group Commercial Lines business (AmTrust Financial – AFSI), and Personal Lines business – (National General Holdings Corp -?NGHC, which recently went public).

With renewal rights, the AFSI & NGHC acquire the assets and the right to renew existing business at terms mutually acceptable to clients & companies, but they do not acquire anything that pertains to claims from business existing prior to the deal. ?In return, they pay money to ACP Re, Karkunkel’s private company owned by his grantor trust.

On question 2, the answer is not likely. ?When Argonaut bought out PXRE in a reverse merger, the NOLs were disallowed. ?Now, I’m not a tax expert, so maybe someone reeeeally clever can fox their way around this, but to me the answer is no.

On question 3, the answer is I don’t know. ?The public companies that he controls have the advantage that they aren’t taking on much risk in a renewal rights transaction. ?Whether they are paying the right price or not depends heavily on whether the reserving for claims at the Tower Group entities is overstated or understated. ?Prior under-reserving may not have been fully corrected. ?With smaller companies near bankruptcy, like Tower Group, there is the risk of death via many cuts.

That brings me to my main insight. ?Though there are no financing contingencies to this deal, ACP Re can walk away with no penalty if it merely wants to do so. ?If they find a material adverse change, the deal can die, and TWGP will have to pay ACP Re a breakup fee.

Like Fairfax Financial’s offer to buy Blackberry, Prem Watsa had the equivalent of a “free look.” ?Tower Group is desperate enough that they gave a “free look” to the Karfunkels and their allied companies. ?The deal is not a lock, and a lot depends on what is written when the late 10-K is finally filed.

Why delay the 10-K? ?My best guess is trying to get the claim reserves right. ?After having to revise reserves twice before, the odds of further revisions are significant. ?You have to understand that claim reserves for P&C companies are not a science, particularly for long-tailed lines, and Tower Group was overly aggressive in those lines.

But delay in filing the 10-K is not a positive sign. ?If you have confidence in the actuarial analysis of reserves, why delay the filing? ?Every other aspect of a P&C insurance company can be calculated within a few weeks of the year’s end. ?No mysteries, except for the reserves.

So, if ACP Re concludes that the likely claim payments from the legacy business are likely to be larger than the net amount they are paying for the legacy business ($67 million), ACP Re can walk away, with no breakup fee. ?In that scenario Tower Group could head to bankruptcy.

So, when I consider the arbitrage opportunities available by buying Tower Group common stock, I would pass. ?As a rule, I don’t short, though I would be tempted to do so here. ?Tower Group is a very complex company for its size, and as such, I have less confidence in its financials. ?Complexity in financial companies creates inflexibility, which can lead to trouble when regulators deny moving cash from one company to another, which might lead to default on debts.

Avoid this situation, and all of the companies involved. ?Buffett has his “Too Hard” pile. ?This one is too hard, because no one can know the claims that will be paid from aggressively written legacy business.

Full disclosure: no positions in the companies mentioned

From respected reader:

Just did a quick calc based on NWLI earnings and thought I would pass onto you as I know you at least used to hold it as a double weight. ?Let me know if you think there are major holes in this theory:

From the Annual Report:

“The yield on debt security purchases to fund insurance operations rebounded somewhat to 3.53% in 2013 from 3.37% in 2012 but was still below the 4.18% yield attained in 2011.”

So, investment yields improved, but are still down. ?Their unrealized gains in securities dropped from $541 million to $146 because of this, so this part of the “hidden value” in the shares went down.

But if rates can get back up to that 4.18%, a quick calc says that would cause annual earnings on their $9 billion investment portfolio to increase $58 million. ?If 2/3’s of this is credited to annuity holders, it leaves $19.5 million before tax for shareholders. ?32% tax from 2013 gives after tax earnings increase of $13 million or $3.57 increase in earnings per share.

If we could get yields back up to 5.5% like they were a few years ago, using the same calc would give an increase in EPS of $9.38, or a 1/3 increase in earnings.

It is still a double-weight here. ?It is not as cheap as it once was, but it is still cheap. ?Financial stocks should always be valued on a combination of price-to-book and price-to-expected-earnings.

Why? ?Because accrual items in the accounting can either be aggressive, fair or conservative. ?If aggressive, earnings will be overstated, and book value understated. ?If conservative, earnings will be understated, and book value overstated. ?For the most part, the two measures balance the squishy accounting.

Now as for the disclosures in the NWLI 10K, we need to note that more than 2/3rds of the bonds that they hold are “held to maturity.” ?That’s unusual, as is their policy where they don’t buy high yield bonds. ?Held to maturity means the value of the bonds amortizes over time, but price moves don’t affect the accounting, unless default is likely. ?Thus if interest rates rise, book value will not be affected much, but earnings will rise on a GAAP basis.

NWLI has a conservative investing culture, and in the present aggressive environment that is a *good thing.* ?Adjusting for the held to maturity securities, the adjusted price-to-book is 55%, and my estimate of future earnings is one-ninth of the current price. ?It is rare to find stocks trading at a significant discount to book and a single-digit P/E.

Full disclosure: long NWLI