Well, thank you Rolfe Winkler and Reuters.? I go off-line on Sundays because it is the Sabbath, so I don’t review the web or catch e-mail, but when Rolfe e-mailed me and I saw it on Monday morning, I felt I had to give him a response.
Now, my response was a brief one, for me.? There was more to say, some of it of a personal nature, but I was busy this weekend.? We moved six of our children into different rooms, repainted two of them, and simplified our lives — I have more trash sitting outside than I have ever seen in my life.? The house is simpler and prettier than ever before, and our two oldest now both have their own rooms.
So, what I wrote was significant, but limited.? Let me fill in some gaps.
First, the efforts on life settlements have been going on for a long time.? This is not new, but has been happening for a little less than 20 years.? Over the last ten years I have been personally invited to be a part of three (maybe more) of these enterprises, and I have turned them all down early because of the ethical issues involved.? I genuinely believe in the concept of insurable interest.
Second, life insurance is for the most part sold, not bought.? I used to have trouble with that, but there are many people who will not save or seek protection unless someone goads them to do so.? Those who will not actively look out for themselves pay a price relative to those who seek coverage unbidden.
Surrender charges exist on life insurance policies to allow insurers to recover the cost of the commission that they have not amortized.? As GAAP accounting would suggest, all revenues and expenses are spread over the life of the policies.? The significant cost of acquiring a life insurance policygets recovered over the life of the policy.? If a policy owner wants to surrender early, the insurance company has a surrender value or cash value that reflects no loss to the insurer on average.
Pretend for a moment that you are a life insurer.? You want to make a profit, or if a mutual, break even.? You test/underwrite potential insured lives before the policy is issued to assign policies to the proper rating class for them.? The more accurate you are, the more polices you will write, and the fewer surrenders you will have.? But over time, people change.? After issue, insureds tend to get more sloppy in their lives, on average.? Also, things that could never have been caught in underwriting emerge.
Those doing life settlements aim at the policies where there have been negative health events since issue.? Death is considerably more likely, and so the value of the policy is worth more.
Think about it: you as the insurance company did your best job to estimate the risk involved. You did it assuming that policies could not be sold, whether really or synthetically.? You already knew that those who were healthy in the future would surrender and seek another carrier, but thought the those who were less healthy would persist to some degree.? Well, with life settlements, the unhealthy persist at a much higher level, which bites into profits.
This is the box that life insurers are in.? They can’t lock in policyholders, but policyholders can hang on, refinance (so to speak), or sell off their obligations.? That is a tough equation for life insurers to work through, and to the degree that life settlements are allowed, premiums will have to rise to compensate for the loss of profitability.
David — thank you for a common sense explanation of life settlements. My gut told me it sounded too good to be true, but I could never explain why. Now I get it.
Wonder if I (we) could pick your brain on another “free lunch” problem: GNMA bonds carry the full faith and credit of the US government, just like Tresasuries. The notable difference is that GNMA’s are subject to prepayments; but in practice the sorts of mortgagees that qualified for FHA loans have limited means to make prepayments even when it was advantageous to do so. The prepayment risk was/is there, but less than “mainstream” mortgages.
So why are GNMA bonds trading at such a massive discount to Treasuries? I looked at several GNMA-only mutual funds (since they are public and no one can discuss private portfolios here). The yields are 230bp or more above similar duration Treasuries — way more than the embedded call option should be worth.
Some of the difference is liquidity, but even if one adds prepayment and liquidity risk — the spread is very wide.
With that preamble — it seems like the market is pricing in a lot of “credit” risk (maybe it should be called political risk). The “full faith and credit” of the US government isn’t quite as full faith when said government is already financially over extended. The headline risk (to politicians) of defaulting on GNMA bonds is a lot less than defaulting on Treasuries
The US markets have never had to deal with pricing US sovereign risk — but after watching Iceland implode, seems like we should be at least thinking about it.
Everyone was complacent about buying AAA securities a year or two back — is the “full faith and credit of the US government” the new AAA rating?
PS — a similar argument applies to muni-bonds, which are rough equivalents to GSEs like FNMA and FHLMC. The US government has no legal obligation to bail them out, but its impossible to imagine California going bankrupt without effecting Uncle Sam.
How do you measure / control the political risk?
Gary, technically GNMA’s are not backed by the gvt. In practice, we know what happened. Banks used to make that trade back when they had balance sheet; that’s gone away and now the spreads have widened. Also, there is a lot of flight to treasuries, so that pushes bond yield down.
David, one thing I don’t get. Why is there any negative effect on the insurance company? They didn’t misprice the risk. They issued a pool, and some people will die prematurely. The trade is the insured saying that he will sell this policy. It’s like someone selling a “winning” lottery ticket. The lottery system is unaffected.
JoshK — that is not true. Unlike FNMA and FHLMC, which only carried an “implied” backing — GNMA bonds have an explicit and official backing. They are 100% backed by the US government; no ifs, ands or buts.
Its amazing how many Wall Street folks trade GSE bonds and don’t bother to read up on them. FNMA and FHLMC were never explicitly backed, no matter how many “analysts” said otherwise. FNMA and FHLMC were investor owned, and legally only had the right to borrow $3 billion apiece from the US treasury.
GNMA bonds were and are explicitly backed. GNMA is 100% owned by the US government. GNMA backs bonds – it is (or was) an insurer like AIG. But the government backs GNMA, explicitly and without any statutory limits. Go look at the website if you aren’t sure.
As for the flight to Treasuries, I agree — but therein lies the problem. GNMA’s are Treasuries (at least from a credit risk prospective) with prepay risk.
Some of the liquidity risk of GNMA’s no doubt stems from Wall Street’s ignorance of what does and does not have full government backing
@ Josh: As the Gary said, “GNMA bonds carry the full faith and credit of the US government, just like Treasuries.” Period. Perhaps you are confusing them with GSE bonds (Fannie, Freddie), which are not full failth and credit issues, despite whatever disinformation Ben Bernanke puts out suggesting otherwise. If it ain’t in the statute, it ain’t guaranteed. That said the “implicit” guarantee is almost airtight for GSE bonds; the same can’t be said for munis, and in fact, the Fed Chairman has said (like any true Republican party hack), that there would be Federalism-based legal impediments to any bailout of the state based muni market. Apparently, Federalism doesn’t bar bailouts of private entities that own the Fed, eg., Goldman Sachs.
David,
While I appreciate your thoughts on the subject, I would have hoped you would provide a little in the way of full disclosure. If life insurance is something that should be sold, do you sell these products?
Do you provide transparency to your clients with regard to commissions, fees and expenses, not by a 700-page prospectus but verbally? Do you demonstrate the option of term life with the substantial difference in premium invested in a no-load government bond fund or ETF?
Despite the risk undertaken by the life insurers, the permanent life products they sell are designed to fail the policyholder. After retirement, the need to protect income disappears, but the premiums remain. For most, negative cash flow investments and retirees go together like peanut butter and mayonnaise. By any estimate, north of 80% of these products will never pay a death benefit, cashed in for a return that would rival that of a simple savings account.
Wall Street is not creating demand where it does not exist. It’s there and it’s real. And as long as the Fed keeps fixed income returns to the bare minimum, Wall Street’s clients, many of which will be life insurers themselves, will see life settlements as a important vehicle to help meet their future obligations, to say nothing of helping what could be millions of seniors repair their impaired balance sheets.
Mike — full disclosure: I am a life actuary who has been a critic of my industry for the last 20 years. I have consistently encouraged people to buy term unless they have a complex tax planning need, or if they are very sick, in which case permanent insurance can be valid.
I derive no income from life insurance or annuity sales, and have not for the past 11 years. Annuities? I have been a critic there as well, but have derived income from the area, though I have never received a sales-based payment.
For some of my criticisms please consult the insurance articles listed on this page:
http://alephblog.com/major-article-list/
To answer another question, on how this hurts the insurance company: most pricing actuaries price in a certain amount of baseline surrender activity when it is not in the interest of the insured to do so. When surrenders drop in later years, the present value of total claims rises more than any incremental increase in investment income.
Oh yeah, sorry, I didn’t read closely. I was thinking about the other GSE’s bond issues, but I think the same thing is true there, just no balance sheet to arb it. But I’m not sure why a foreign buyer would pass them up for lower priced treasuries. Let me ask our rates desk head.
it seems to me that Bernanke has entered a Faustian bargain with the primary dealers: “The Fed will pretend you guys are still solvent if you pretend US Treasuries are still AAA”
Part of that bargain is the dealers must buy lots of Treasury bonds using the Fed Funds they get at 0%. Foreign CBs seem to be buying *less* (not zero, and they aren’t selling). Joe Q America can’t pay his mortgage and his 401K has been trashed — so a sudden surge of Treasury buying with money he doesn’t have seems unlikely. Its Ben Bernanke buying Treasuries — albeit using banks and primary dealers as front men
Defaulting (broadly defined) on GNMA bonds carries a lot less headline / political risk for the Treasury than defaulting on Treasury bonds. The liabilities could be “restructured” with longer maturities and/or lower coupons — which is still a default but something the news weenies could spin other ways.
The reason I am asking all this about GNMA is that Congress seems to be pushing lower and lower quality lending onto GNMA’s books — very similar to what they did to FNMA and FHLMC.
The Treasury can’t “bail out” GNMA, since they are two parts of the same whole (Uncle Sam) — but any failure would be even more spending of money the government doesn’t have.
And since David seems to be able to articulate things in common sense terms, my big question that started all this:
how does he model political / sovereign risk?
@maynardGKeynes: you are confused. Goldman Sachs owns the US Treasury (or thinks it does) 🙂
As for bailing out the muni market– there was no legal basis for bailing out AIG (which was really a GS bailout in drag). That was Henry Paulson looking after his cronies.
Bernanke is right that there are legal problems in bailing out muni bonds — but in the end it all comes back to asking how an insolvent Federal Government can realistically bail out any more entities of any kind.
Inflation / USD depreciation is the obvious political answer — but even super model Gisele noted that currency devaluing didn’t really help the banana republics of the past
I worry that GNMA is too big to save — much like Citi, BofA, etc. So the powers that be will have to torture the accounting code even more.
And I don’t know any good way to model / think about how that might unfold…
David?
David, isn’t this just making the market more efficient then? Essentially, the insurance company priced the policies as if no one besides themselves could bid on them. If someone had cancer and wanted to get what cash they could from their policy then they had few options. Now the insured has more options and can make a better cash-out when sick.
David,
Thank you for your response. I will read your take on the industry and its’ products via the link you provided. I agree that permanent life has its place in certain circumstances, but like interest-only or neg-am mortgages, the products were eventually pushed on the public as a whole by an industry that could not look beyond the next quarter and their representatives that could not look beyond the commission.
I think it is important to make the distinction, which you did, between term and permanent. And it is equally important to note that investors are only interested in policies of the permanent variety.
You are correct that this is not a zero sum game – insurers will see fewer policies redeemed than expected should the settlement industry expand with or without the assistance of securitization. So they will be the losers, but protected by the Too Big to Fail mandate. America’s seniors, on the other hand, have no such protection and any attempt to “protect” them from Wall Street will have the predictable opposite than intended effect.
side note, I have long wondered why no one offers a combination life – health insurance policy.