Description

Mercury General was founded in 1961 by 40 year-old actuary and life insurance salesman George Joseph, who raised $2mm in capital and sold his first auto policy in California in 1962.  Sixty years later, Mercury is now the 7th largest writer of personal auto insurance in California, and the 16th largest overall in the country.  It writes about 1.8  million policies annually, mainly private auto insurance (~1 million policies), homeowners (~760k), and commercial auto (~40k).  

And George is still kicking at 102 and owns 35% of the company -- worth about $900m -- while his (much younger -- she's 96) wife owns another 16.5%. Together George and Gloria control the company with George as chairman. 

The Josephs have held on to their stock for decades and as far as we can tell the only sold a small amount once three years ago when the stock at $65 -- otherwise, the Josephs have never sold a share (in fact he has been a net buyer).  As chairman, George's compensation is around $1.5 million a year (modest, given with his control, he can pay himself almost anything).  The Joseph family seems to put food on the table almost entirely with their dividends which they are paid alongside the outside public shareholders.  Historically, Mercury has paid out prodigious dividends -- prior to 2022, dividends per share have grown at a compound annual growth rate of about 7-8% over 30 years, depending on the start/end dates.  

Another quirky -- perhaps quaint -- feature of this company is that shares outstanding don't really budge -- the company's frugality with equity awards is a rarity in today's world -- share creep has averaged less than 0.1% over the past decade. In fact, the company hasn't issued any options or RSUs to management in five years, and then only negligibly.  Share count has stood at around 55 million for many years.  Don't expect a buyback either - they haven't done one in 24 years.  Internally generated cash flow (and there is a lot of it) and excess capital are plowed back to support writing more business, and otherwise paid out as dividends ($1.3 billion in total over the last decade -- roughly 50% of the current market cap). 

The people in this story are important because the business itself is not exactly attractive.  Private auto insurance is commoditized, heavily regulated by finicky state insurance departments, and generally not exactly a fun purchase for consumers.  It is a low-margin, price competitive business.  On top of this, Mercury operates mainly in the great state of California -- it's 79% of Mercury's premiums -- a notoriously difficult insurance market due both to fickle (anti-business) regulators as well as a propensity for crazy natural disaters - mudslides, wildfires, storms, et cetera -- many of which destroy cars and homes insured by Mercury.  Meanwhile, Mercury competes with massively larger, lower-cost, better-known carrieres like Progressive, GEICO and Allstate.  Not fun. 

The upshot?  Over the last 20 years, MCY shares are slightly down --- total return has been a measly 4.4% annually, with dividends.  

Ouch.

Still reading? It gets worse.  Last year, the company wrote $4.3 billion in premiums and earned $16 million.  That's on a book value of $1.5bn -- so a 1% ROE for those keeping score.  Value destroyed - check.  If you think that's bad, know that it was a marked improvement over 2022, when Mercury lost $127 million in and incinerated $618 million of book value -- the annus horribilis for car insurers nationwide.  There are a couple of VIC writeups on PGR and ALL from 2022 and 2023 - these may provide background on what happend in 2022 (and extened into 2023) in the personal auto insurance business.  Simply put -- it was BAD, a perfect storm: spiking cost inflation for car repair and replacement and bodily injury medical costs destroyed loss picks across the industry. And because all or much of the business is written on an 'admitted' basis -- in regulated markets -- the carriers could not apply for, get permission to, and enact prices increases nearly fast enough.   The result was wholesale hemorraging of money by all carriers, even the good ones and irrespective of geography or scale: GEICO and Allstate (auto) both lost 4-5 cents for every dollar of premium written that year (for ~104% combined ratio).  Don't worry, Mercury fumbled it even more badly, it reported its biggest operating loss ever in dollars and worst combined ratio -- at 108.7% -- in at least 30 years.  

So what gives -- why are Mercury shares a buy now?

A shibboleth in cyclical commodity investing is that when the cycle turns -- or if one anticipates a turn -- one ought to buy the highest-costs, most leveraged, lowest-quality player in the sector which has the most explosive upside to profits when conditions improve.  Get leverage to the cycle.  In March 2009, you wanted to buy BAC, not JPM.  You get the point:d on't buy quality; buy junk.  The junk with the most operating and financial leverage and the lowest starting point for profits.  

Well, I don't believe that Mercury is junk.  I think highly of Mercury.  The company has navigated the trecherous California insurance market rather admirably for a long time, has protected millions of policyholders, has managed to stay independent, pump out dividends, and be (mostly) profitable for six decades running.  That's impressive! 

But the business is hugely, unusually, cyclical, even for a personal auto insurer, by virtue of its concentrated exposure to one market (California), and to a lesser degree, its small size with high fixed overheads relative to its volume of business.  In just the past four years, MCY has earned as much as $5.54 per share (in 2020) and has LOST has much as $2.30 per share in 2022 (all numbers are on an operating basis, excluding large swings in realized and unrealized investment gains and losses).  Underwriting margins (1 minus the combined ratio) have been as good as positive 6.9% in 2020 and as bad as negative 8.7% in 2022.  The aforementioned catastrophe events add some much-unappreciated volatility to results:  the company paid out as little as $53m in catastrophe-related losses back in 2019, but that figure jumped to a whopping $239 million in 2023 (biblical stuff: rainstorms in TX, hail in Oklahoma, tropical storm Hilary in CA). 

As 2022 progressed, auto insurers began to realize they were in a bind:  people were driving more, and inflation was out of control.  Used car values increased by over 60% between 2020 and 2022, labor costs shot up, supply chains were messed up, and aftermarket parts were not available.  Replacing or repairing cars got more expensive - A LOT more expensive.  As MCY's wrote in its 10K, "in 2022, inflationary trends accelerated to their highest level since the 1980s."  Mercury and its peers rushed out to raise prices but in the admitted (state regulated) market, the Department of Insurance (or similar) needs to approve every rate action.  This process works on government time.  A loooooong time.   State insurance regulators are programmed to push back on carriers and to try to protect consumers (or at least give the outward appearance of protecting consumers).  That meant that rate approvals came very slowly and not in time.  The industry barely got rate increases in 2022 (which resulted in MCY losing $344 million underwriting that year).  Mercury cut its dividend in half in August 2022 because it was at risk of running out of capital if price increases were not forthcoming.

This is the point in the story when the clouds start to part.  State regulators began approving meaningful rate hikes in early 2023.  It was kind of too little and too late to salvage 2023.  Mercury was able to increase policy prices, on average, about 14% in 2023 -- not really even enough to cover the loss cost inflation from 2022, not to mention persistent and compounded loss cost inflation (though moderating) in 2023.  But, as often the case, the regulators over-compensated, perhaps realizing how far behind the carriers were on rate vs loss costs (maybe the regulators panicked, too, after Allstate stopped writing new policies in June 2023).  This over-compensation on rate actions, combined with moderation of loss cost inflation (as supply chains untangled and used car values fell), meant carriers started making money again.  

For MCY specifically, it started making money - A LOT of money -- in the third quarter of 2023.  Not coincidentally, 3Q of 2023 was when it started realizing the full benefits of two approved rate actions -- the first in March of 2023, and the second in July of 2023.  By 3Q23, Mercury's pricing was running around +21% Y/Y, a monumental improvement over the preceeding quarters.  In 4Q22, 1Q23, and 2Q23, premiums-per-policy (proxy for pricing), was DOWN -2%, up 1%, and up 13%.  Commensurately, operating EPS improved from a LOSS of $0.47/share in the second quarer to $1.14 in the third quarter.  In 4Q23, operating EPS further improved to $1.15.  

With loss costs relatively stable and the its book of business properly priced, MCY is annualizing to an earnings power of around $4.60, based on 3Q and 4Q23's EPS of $1.15 multipled by 4.  That's simple enough.  3Q and 4Q included some catastrophe losses that were in the range of normal (**about $100 million annualized.  For the time being, let's assume $100 million in annual cat losses is "normal" -- this will be a swing factor (sometimes significant) from year to year, but not one that affects the thesis in any meaningful way.  MCY purchases a fair amount of cat reinsurance treaty with a $1.1 billion limit, which essentially limits losses on individual events to around $80 million; as a CA writer, MCY lays of earthquakes risks on its homeowners policies to the California Earthquake Authority).  

So we have an earnings base of $4.60 as our starting point.  Last year's rate actions will continue to flow in through this coming July, when they fully anniversary. Meanwhile, loss costs do appear stable:  losses per policy-in-force (excluding cat losses) were inflating at 15-16% throughout 2021 and 2022, but increased only 5.6% in 2023 with fourth quarter losses/policy actually down -1% Y/Y.  I wouldn't bet on cost deflation quite yet, but the trend is encouraging.  This should provide some comfort that the aforementioned $4.60 is a decent proxy for earnings power for 2024, maybe 2025.  

This is when things get really interesting.  The California regulator this past January approved a 22.5% rate increase for the personal auto line of business which account for nearly 50% of MCY's total premiums written.  This rate increase became effective in late February of this year.  This is potentially the largest rate increase in MCY's history -- definitely since at least the 1980s. 

Apparently (and i speculate), the California DOI did not want to see the state's auto insurance market becomes dysfunctional -- already, Kemper (which is the 13th largest auto insurer in CA) announced it is exiting from the state. Last year both Allstate and State Farm made various threats to pick up and leave.  GEICO is the 3rd largest auto insurer in CA; they don't like losing money for long.  Imagine being the insurance regulator under whose watch citizens lost access to federally-mandated car insurance.  Not going to happen.  Regulators are anti-business -- to a point.  That point was reached sometime last year.  I think the CA DOI realized they have to go above and beyond to ensure the future viaiblity of car insurance in the state, by giving the carriers an opportunity to repair their balance sheets (build capital), and in doing so -- allow them to over-earn for a while. Of the pure play public personal line insurance companies (of which there are few), MCY is by far the most levered to this dynamic, and stands to benefit the most.  

As the February rate action flows through to Mercury's business, it compounds upon a 6.9% rate increase taken in March 2023 and a further 7% increase taken in July 2023.  Cumulative price increases of 40% since the beginning of 2023, yes.  That might go a long way to raise margins and repair balance sheets.  Mercury is currently anniversaring the March-23 increase, but pricing overall we estimate should be about +18% in 1Q24, +15% in 2Q24, and around +10% in the second half of the year (as prior rate hikes anniversary, and assuming no further approved rate actions).  For 2024, we think MCY's realized pricing should be in the rang eof +15-20%, with bias to the high end of that.  Take these numbers with a small grain of salt as they are estimates -- but we think pretty close.  These price actions are coming at the exact time when loss costs are moderating, which should result in super-profits for Mercury.  Temporary perhaps, but super nonetheless.  

Nothing on the horizon should leave one with the impression that loss costs are about to reinflate.  In the 15 years prior to the pandemic year, loss costs ran at around 3% per year. We are assuming 4% in our model.  With pricing running around +15%, we think combined ratio should end up around 94% in 2024 -- for a 6% underwriting margin.  This includes a reserve for normal cat losses.  In 2020, MCY did a 93% combined.  Prior to that, MCY's best years were in the 2003-2007 time frame when the combined ratio ranged between 89% and 95%.  All this to say -- we don't think a 94% CR assumption (or as it were, output) is unreasonable.  

On the current book of business a 6% underwriting margin is $300m in pretax profits.  Add around $250m in investment income and earnings power of $550m equates to $8.00 per share.  Now, earnings power is a matter of enormous leverage; every 1 point of combined ratio is about $0.70 per share to earnings -- a nealry 10% swing.  So we might be right -- or we might be very right, or very wrong.  We are biased toward being very right and too low on the underwriting margin.  If cats are mild, the number will be north of $8.00, and vice versa.  

In any event, we think a starting point for earnings power for Mercury is $8.00 for 2024.  Shares are currently at 6x this number.  Is this multiple too low?  Maybe, maybe not. For many years in the mid-2000s, the stock traded at a single-digit p/e, but it also traded as high as 15x at times.  Book value should increase to around $35 by year-end 2024 so we're around 1.4x. Again, what is the right multiple?  We don't know, and comps like PGR and ALL (which trade at 2.6x - 5.5x book) are not appropriate.  We do think that if MCY earns $8 this year, the stock will be a lot higher.  For two reasons.

The first reason is capital returns.  As the company rebuilds its balance sheet, it is likely to double its dividend to the pre-2022 rate (which would be a 5% yield).  If the company can sustain the super-profit profitability out to 2025 (we think it can and model $8.00 in 2025 as well), MCY will be sitting with excess capital of about $300mm -- around $5/share.  If history is guide, it will dividend this back to shareholders (with the Joseph family getting 52%).  The farther out we look the murkier MCY's earnings picture looks:  will state regulators claw back the rate actions?; will there be consumer backlash? (most likely - yes); will 20%-plus ROEs attract new entrants to the market (obviously yes).  This is not a business that will print 20% ROEs forever -- or even for very long.  But it is a business that we think could earn 30-40% of its current market cap in the next two years.  

The second reason we think the stock will be a lot higher comes full circle to 102 year old George Joseph.  As best we can tell, the Joseph family shares are held directly by Mr. Joseph and another tranche is held by his wife; they are not held in a trust.  George's 37-year old son is newly-minted chief operating officer and owns essentially no stock.  We don't like to speculate on stuff like this, so we think suffice it to say that a very large estate tax liability is likely to be triggered at some point in the coming years.  If the California personal lines market indeed fundamentally heals, the state could become attractive to aspiring entrants and Mercury is probably the most logical way for a big carrier to buy its way in.  Past acquisitions were done at 1.5x to 2.7x book which we think puts the private-market value at around $70-75 per share.  An earnings-based takeout valuation suggets something closer to $80-85.  

Mercury is not a good business.  ROE has exceeded 20% in just five of the last 20 years.  The company isn't growing -- its book of business has grown just 12% in the last 20 years.  But it's fairly simple to understand, and it's a unique and interesting time to invest in this business, ahead of what might be a 'generational' hard market in its key markets.  The stock has done nothing in years and is a bit of a relic (one sellside firm covers it; they stopped doing earnings calls a few years back).  

Valuation is reasonable (10x) on the current earnings power of $4.60, and very cheap (6x) on where we think earnings power will be for the next couple of years.  A sale of the company is possible given the founder's age.  Mercury has been a longtime rumored takeout target, as far back as 2008 and earlier.  But hey, maybe we'll get lucky.  

Obligatory disclosure:

Don't rely on any of the numbers cited above.  Many of them are our estimates, and some of them are from third-party sources, and none may be all that reliable, if at all.  We own the stock, and are making various assumptions which may not be accurate or may turn out to be wrong.  

Catalyst

Earnings -- as investors realize EPS power is $8, not $3 

Takeover offer?