Insurance

Given Progressive’s adverse loss development in Q1, we investigated whether there is enough profitability and reserve stress to lead to a pricing turn in Personal Auto.  Unfortunately, this does not seem to be likely: while Progressive’s reserves are only barely adequate (1.3% of 2011 premium), most other companies seem reasonably adequate (around 5% of 2011 premium)

Perhaps worse for Progressive, its main rival for growth, GEICO (part of Berkshire Hathaway), has exceptionally strong Auto reserve adequacy at about 17% of 2011 earned premium.  So it seems more likely that Progressive will need to unilaterally strengthen reserves in the near-term, putting its earnings at somewhat of a risk versus peers

All together, this analysis supports our Short-Term Neutral View on P&C Personal: there is not enough stress to lead to a pricing cycle near-term, but neither do we see enormous competition in the offing

As a side benefit, we completed full company and line reserve analyses for the 5 public companies in this review: Berkshire, Progressive, Allstate, Travelers, and Hartford.  Berkshire’s overall reserves are as exceptional as its Auto book, about 18 points redundant, with no individual lines significantly deficient.  The other 4 companies have seen reserve deterioration over the past 4 years, and now are all within the statistical range of no redundancy or deficiency

In our 2010 US industry loss reserve analysis, we saw many indications that reserve adequacy was holding up better than expected.  That completely changed in 2011: reserve adequacy deteriorated across the board in Commercial lines, and only improved for Personal Auto.  While still not at the 1999-2000 levels of deficiency, the pricing action seen in 2011 now makes more sense

We estimate that 2011 US industry loss reserves declined to $10.6 billion from $19.1 billion at year-end 2010 (Exhibit 1).  This is only about 2% of 2011 earned premium.  The standard deviation of this estimate is about $10 billion.  Significantly, we estimate that this is the first year since 2002 that the current accident year has been booked deficiently in total (by about $600 million)

The overall estimate masks much deterioration in the core Commercial lines: Small Commercial and Professional Indemnity are deficient (12% and 11% of 2011 earned premium, respectively), and General Liability and Work Comp adequacy fell by large amounts ($4bn and $3.4bn, respectively)

There are still cross-currents, but they are mixed: our estimate last year that Work Comp trends were staying moderate did NOT hold up in 2011, but did for General Liability, though current paid and reported emergence is getting worse here, for the first time in a long while

This analysis supports our Short-Term Positive View on a commercial pricing cycle in the “usual” way, though likely more muted given that the industry is reacting faster than historically.  This applies to Brokers, P&C Multiline, P&C Commercial, and P&C Specialty

Despite our current Short-Term Positive View on subsectors exposed to commercial lines pricing improvement, we are Long-Term Negative on most insurance underwriters, including Life.  Our overall thesis is that insurers have written too much risk that the market does not want (e.g. annuities, long-tail), cannot easily find ways to write risk the market does want (e.g. tradable), and will need another set of “crises” before it is driven to make any major changes

Today’s note focuses first on the lack of growth relative to the overall economy (Exhibit 1).  For the last 25 years, auto and liability insurance have been shrinking as a percentage of GDP, as have mortality and annuity products for the last 10 years.  Only Life A&H appears to be a major insurance line that is growing with a healthy pricing cycle.  While property insurance appears stable versus GDP, the details reveal that it has mostly been raising price and cutting exposure

While insurance investors and insiders understand that not all growth is good in insurance (i.e. profits and risk matter), this is a much smaller issue for equities outside of financials.  Without some growth, it is difficult for insurance to be seen as a vehicle for outperformance, except for occasional periods of insurance price correction in P&C, or market events in Life

When we strip out price from exposure, we see no growth versus GDP at all, except for A&H and cyclicality.  Worse, in Life insurance, we do see continued growth in reserves, so Life insurers are accumulating older “obsolete” exposure while newer “better” exposure declines

The subsectors where we are Long-Term Positive are areas where we see the potential for innovation to ignite future growth (P&C Personal, Life Other), or where the imperative to aggressively manage older exposures is already present (Multiline)

Going into Q1 earnings, we provide investors with this summary of our recent research and key investment themes to watch.  We provide a “tear sheet” for each of our 11 insurance subsectors summarizing our key conclusions

We are changing our Long-Term View on Multiline to Positive from Negative.  While we still think the pure multiline model is a dying one, we now think that this is sufficiently evident that companies will start making the needed changes.  Companies like AIG and HIG are in the midst of this process, and while we expect short-term uncertainty (i.e. View remains Neutral), particularly of a regulatory nature, investors should start looking at the remaining US multilines for break-up and selective restructuring opportunities

Better equity market performance and less risk aversions means our Short-Term Views have largely underperformed.  Insurance overall underperformed 3% in Q1, and the subsectors where we hold Short-Term Positive Views underperformed even more (-6% to -10%).  Conversely, Life Investment was the only subsector to outperform (12%) by any significance.  We remain cautious about risk aversion “returning to normal” near term, but we acknowledge that any such normalization is the single biggest risk to our overall Short-Term Qualitative Views (i.e. we may be “fighting the last battle”).  We are not including any kind of market forecast in our analysis, but for investor bullish on equities overall, we think that would be bad for our Positive View on P&C and Brokers, and more bullish for Life Insurance

Investors seem appropriately skeptical of the nascent commercial lines pricing cycle.  Companies continue to raise prices seemingly in the absence of severe balance sheet problems.  Depressed profits, including low investment yields, has seldom been sufficient to sustain historical pricing, yet the evidence for accelerating and broadening pricing (e.g. from property to work comp to professional lines) continues to mount.  As a result, playing commercial-exposed names (Brokers, P&C Multiline, P&C Commercial, P&C Specialty) remains our primary Short-Term View, despite underperformance over the last 6 monthsThe following recent research is available for more detailed discussions of these issues:

A Portfolio Manager’s Guide to Global Risk Management & Insurance (December 7 2011): Detailed analysis of short and long-term stock performance by insurance subsectors, designed to provide portfolio managers with a basic understanding of insurance stock behavior

Insurance in 2012: It’s (Still) About the Risk (January 3 2012): An analysis of the effects of heightened risk aversion on insurance subsector performance, which we continue to believe will be an important factor in stock selection, despite an easing of risk aversion across the market in Q1

P&C Pricing Expectations: Backing Off Reinsurance, Play It Through Specialty (January 22 2012): We examine the prospects for future commercial and reinsurance pricing in the context of historical cycles, and conclude that specialty companies have often been a better way to play reinsurance

Life Insurance in 2012: Jumping a Big ROE Hurdle (February 16 2012): We show that post-crisis valuation declines in life insurers have contributions from both earnings challenges (lower yields & leverage) and higher equity risk premia, which present a big hurdle for improved long-term stock performance

Personal Auto Game Change: The Future May Not Be The Past (March 28 2012): Personal auto insurance, historically one of the better insurance subsectors, is facing pressure from a closing performance gap with non-public competitors.  This suggests that innovation will need to be an ever-increasing part of the investment case, and we do think this will happen over time

Short-term, the historical playbook of a sharp upward pricing cycle and continued market share gains to stock personal auto insurers seems less likely: These two events are the typical incentives for value and growth investors (respectively) to invest in the industry, and seem more challenged going forward.  A standard cycle seems less likely because the mistakes that tend to drive pricing cycles (e.g. reserve adequacy, loss trend) are becoming less likely for auto insurance.  Persistent share gains made by stock insurers from 1996-2006 have slowed because their ability to out-underwrite mutuals has seemingly decreased. (Exhibit 1).  We cannot permanently rule out either event reoccurring, but the probability of a return to past strength does seem reduced

The stock insurer performance gap over large mutuals is closing, even as it persists versus smaller insurers and State Farm: It is not just that the return on premium margin gap has closed (3-8 point stock advantage over mutuals 2001-07 vs. 0-3 points currently), but also that the gap is now within the range of loss ratio dispersion of stock companies, meaning that implicit differences in the portfolios of stock vs. mutuals is converging

Efficiency gains, which are a core thesis for many auto insurance investors, have proven elusive: Many investors see efficiency as a key issue for auto insurers, but as measured by expense ratio, the results are disappointing, as stock companies have barely budged out of a 21-23% range; if anything, mutuals may have a slight advantage in near-term expense ratio improvement, given a long history of gradual deterioration

So is there a reason for longer-term bullishness?  Yes, from the potential for “game changing” technological developments: These seem most likely to come from increases in mass customization: increased ability to tailor policy terms, which also increases transparency; greater use of seasonality to better tie usage with claims frequency; real-time feedback of all aspects of auto insurance (purchase, claims, repair) to increase the sense of control

But like efficiency gains, technological change may prove elusive, so the stock potential here presents itself as optionality rather than inevitability: Technological advancement in insurance tends to occur more slowly than many expect/hope; impediments include state consumer regulation, lack of standardization (though this is better in auto insurance), and lack of interest on the part of consumers

Progressive (PGR) still seems the best option on developing the next great auto insurance breakthrough: Progressive has led the industry historically (credit-based pricing, concierge auto service) and continues today (Snapshot usage-based pricing, Name Your Own Price policy customization); we suspect that both these technologies have greater future potential as more data becomes available: seasonal premium adjustments (like monthly utility bill); better understanding of policyholder behavior increases value of policy customization; opt-in for more real-time monitoring brings bigger discounts and more data credibility

Allstate (ALL) present the option for favorable disruption of its own business model, with the obvious risks of failure: Allstate has an enormous opportunity to shift its culture in a positive way via its eSurance acquisition; while the track record of such deals is not good (i.e. the acquirer tends to absorb the acquired), nurturing eSurance to the potential detriment of its existing businesses could prove enormously powerful in figuring out exactly what works at Allstate and what does not

Our stance on P&C underwriters in 2012 is decidedly “bipolar”, with a Short-Term Positive View on most commercial P&C subsectors, but a Negative View Long-Term.  This contrast all hinges upon assessment of the evolving commercial lines pricing cycle

Short-term, there are reasons to be positive beyond the optimistic management commentary and renewal disclosures: 1) premium levels are depressed versus GDP in a way comparable to past bottoms (1984, 2000); and 2) reported underwriting returns considering low interest rates are consistent with the premium/GDP metric

But digging deeper, these metrics do appear to be short-term biased: for example, if we smooth catastrophes to long-term averages, 1984 and 2000 still look like cycle bottoms, but 2011 no longer does, better resembling shorter-term cat-driven cycles like 1992

The disconnect seems driven by loss reserve adequacy: we estimated 2010 year-end US loss reserves were $19 billion redundant, and using available 2011 data, year-end 2011 will likely still be redundant by $10 billion.  To be consistent with 1984 and 2000, we’d need more like a $50 billion deficiency (~10% of premium) to get a comparable cycle

Reserve adequacy is a key pricing indicator, as reserve deficiencies almost always result in big price increases about 3 years after the first deficient accident year is booked.  So far, we do not even think 2011 has been booked deficiently, resulting in 2014-15 as our best estimate for when reserve can be deficient enough for a classical turn

Company-level reserve adequacy appears very dispersed, meaning the industry may no longer move together: we examined general liability (GL) at the company level, the most redundant single line and very important to the pricing cycle; several large companies (e.g. CB, TRV in P&C Multiline) appear to have GL adequacy much higher than the average

For the near-term, we would favor subsectors with high perceived safety, like P&C Multiline and Brokers, which have low reserve risk and may even show upside to consensus (i.e. Brokers get top-line leverage to pricing sooner, Multiline reserve adequacy potentially boosts earnings).  P&C Specialty would be a good choice for investors wanting more near-term leverage to the cycle and are willing to take on additional potential balance sheet risk

Interestingly, some analysts and investors are becoming concerned about potentially weaker reserve releases hurting earnings.  We hold the opposite view: there may be upside short-term, but ironically this is what could ultimately derail current pricing, thus driving our more negative long-term view—i.e. a “real” cycle bottom by 2014-15 or so

 Print This Post 

Given the strong start for life insurance stocks in 2012, it is important to re-examine our Short-Term Neutral View on the life insurance subsectors.  In this note, we conclude that the Neutral View remains appropriate, owing to a decline in core return on equity, as well as measurable increases in the cost of equity subsequent to the financial crisis

Using US industry statutory data, we examine the components of return on surplus (RoS) both prior to (2003-07) and subsequent to (2010-11) the financial crisis.  Average RoS has declined 300 bps, driven by sustained lower interest rates (130 bps) and structurally lower operating leverage (160 bps) as companies sought to raise capital and reduce risk.  At least in the short-term, these factors seem difficult to control at the company level

The 300 bps decline in industry RoS is mirrored by a comparable decline in GAAP ROE for mid- to large-cap life insurers.  We use a DCF-type model to estimate the change in discount rate pre- and post-crisis that reconciles ROE with average price/book.  We conclude that life insurers face 390 bps on average in higher cost of capital post-crisis, owing to the higher asset and liability risk revealed in 2008-09

The combined 690 bp net increase in ROE hurdle needed to support valuation represents a high hurdle for investors looking for investment opportunities in life insurance, in our view.  If one holds the view that the apparent increase in life insurer cost of capital is transitory, there may be many more opportunities than we think are available.  We think it would be more prudent to focus on those areas where there is a higher probability to improve ROE

At the company level, Life-Investment “value” plays (e.g. LNC, GNW, HIG) seem more challenged to improve ROE near-term, owing to the stress taken during the crisis.  In contrast, the lower-risk Life-Other names (AFL, AMP) already have post-crisis ROEs above the pre-crisis level.  Neither group seems likely to see near-term decline in risk premium, absent a market event

Some Life Investment names—MET in particular and PFG to a lesser extent—may have a higher probability to increase ROE near-term.  Even a recovery of one-half the post-crisis ROE decline (350 bps for MET, 200 bps for PFG) could result in meaningful price/book improvements (on average 100 bps of ROE or cost of capital equates to 0.1 points of price/book)

Our Long-Term Views on the Life Insurance subsectors remain unchanged.  We remain Long-Term Negative on Life-Investment, owing to insufficient change in the core asset and liability challenges of variable annuities and their associated guaranties, which are not likely to be full addressed absent another crisis.  Conversely, we have a Long-Term Positive View on Life-Other, given its lower risk profile and higher innovative potential, both of which increase of the probability of eventual decline in market risk premium