Still Dancing

The latest market wobble this week comes under the guise of the endless Trump soap opera and the first widespread use of the impeachment word. I doubt it will be the last time we hear that word! The bookies are now offering even odds of impeachment. My guess is that Trump’s biggest stumble will come over some business conflict of interest and/or a re-emergence of proof of his caveman behaviour towards woman. The prospect of a President Pence is unlikely to deeply upset (the non-crazy) republicans or the market. The issue is likely “when not if” and the impact will depend upon whether the republicans still control Congress.

Despite the week’s wobble, the S&P500 is still up over 6% this year. May is always a good month to assess market valuation and revisit the on-going debate on whether historical metrics or forward looking metrics are valid in this low interest rate/elevated profit margin world. Examples of recent posts on this topic include this post one highlighted McKinsey’s work on the changing nature of earnings and this post looked at the impact of technology on profit profiles.

The hedge fund guru Paul Tudor Jones recently stated that a chart of the market’s value relative to US GDP, sometimes called the Buffet indicator as below, should be “terrifying” to central bankers and an indicator that investors are unrealistically valuing future growth in the economy.

click to enlarge

Other historical indicators such as the S&P500 trailing 12 month PE or the PE10 (aka Shiller CAPE) suggest the market is 60% to 75% overvalued (this old post outlines some of the on-going arguments around CAPE).

click to enlarge

So, it was fascinating to see a value investor as respected as Jeremy Grantham of GMO recently issue a piece called “This time seems very very different” stating that “the single largest input to higher margins, though, is likely to be the existence of much lower real interest rates since 1997 combined with higher leverage” and that “pre-1997 real rates averaged 200 bps higher than now and leverage was 25% lower”. Graham argues that low interest rates, relative to historical levels, are here for some time to come due to structural reasons including income inequality and aging populations resulting in more aged savers and less younger spenders. Increased monopoly, political, and brand power in modern business models have, according to Graham, reduced the normal competitive pressures and created a new stickiness in profits that has sustained higher margins.

The ever-cautious John Hussman is disgusted that such a person as Jeremy Grantham would dare join the “this time it’s different” crowd. In a rebuttal piece, Hussman discounts interest rates as the reason for elevated profits (he points out that debt of U.S. corporations as a ratio to revenues is more than double its historical median) and firmly puts the reason down to declining labour compensation as a share of output prices, as illustrated by the Hussman graph below.

click to enlarge

Hussman argues that labour costs and profit margins are in the process of being normalised as the labour market tightens. He states that “valuations are now so obscenely elevated that even an outcome that fluctuates modestly about some new, higher average [profit margin] would easily take the S&P 500 35-40% lower over the completion of the current market cycle”. Hussman favoured valuation metric of the ratio of nonfinancial market capitalization to corporate gross value-added (including estimated foreign revenues), shown below, predicts a rocky road ahead.

click to enlarge

The bulls point to a growing economy and ongoing earnings growth, as illustrated by the S&P figures below on operating EPS projections, particularly in the technology, industrials, energy, healthcare and consumer sectors.

click to enlarge

Taking operating earnings as a valid valuation metric, the S&P figures show that EPS estimates for 2017 and 2018 (with a small haircut increasing in time to discount the consistent over optimism of analyst forward estimates) support the bull argument that current valuations will be justified by earnings growth over the coming quarters, as shown below.

click to enlarge

The IMF Global Financial Stability report from April contains some interesting stuff on risks facing the corporate sector. They highlight that financial risk taking (defined as purchases of financial assets, M&A and shareholder pay-outs) has averaged $940 billion a year over the past three years for S&P 500 firms representing more than half of free corporate cash flow, with the health care and information technology sectors being the biggest culprits. The IMF point to elevated leverage levels, as seen in the graph below, reflective of a mature credit cycle which could end badly if interest rates rise above the historical low levels of recent times.

click to enlarge

The report highlights that debt levels are uneven with particularly exposed sectors being energy, real estate and utilities, as can be seen below.

click to enlarge

The IMF looked beyond the S&P500 to a broader set of nearly 4,000 US firms to show a similar rise in leverage and capability to service debt, as illustrated below.

click to enlarge

Another graph I found interesting from the IMF report was the one below on the level of historical capital expenditure relative to total assets, as below. A possible explanation is the growth in technology driven business models which don’t require large plant & property investments. The IMF report does point out that tax cuts or offshore tax holidays will, based upon past examples, likely result in more financial risk taking actions rather than increased investment.

click to enlarge

I also found a paper referenced in the report on pensions (“Pension Fund Asset Allocation and Liability Discount Rates” by Aleksandar Andonov, Rob Bauer and Martijn Cremers) interesting as I had suspected that low interest rates have encouraged baby boomers to be over-invested in equities relative to historical fixed income allocations. The paper defines risky assets as investments in public equity, alternative assets, and high-yield bonds. The authors state that “a 10% increase in the percentage of retired members of U.S. public pension funds is associated with a 5.93% increase in their allocation to risky assets” and for all other funds “a 10% increase in the percentage of retired members is associated with a 1.67% lower allocation to risky assets”.  The graph below shows public pension higher allocation to risky assets up to 2012. It would be fascinating to see if this trend has continued to today.

click to enlarge

They further conclude that “this increased risk-taking enables more mature U.S. public funds to use higher discount rates, as a 10% increase in their percentage of retired members is associated with a 75 basis point increase in their discount rate” and that “our regulatory incentives hypothesis argues that the GASB guidelines give U.S. public funds an incentive to increase their allocation to risky assets with higher expected returns in order to justify a higher discount rate and report a lower value of liabilities”. The graph below illustrates the stark difference between the US and Europe.

click to enlarge

So, in conclusion, unless Mr Trump does something really stupid (currently around 50:50 in my opinion) like start a war, current valuations can be justified within a +/- 10% range by bulls assuming the possibility of fiscal stimulus and/or tax cuts is still on the table. However, there are cracks in the system and as interest rates start to increase over the medium term, I suspect vulnerabilities will be exposed in the current bull argument. I am happy to take some profits here and have reduced by equity exposure to around 35% of my portfolio to see how things go over the summer (sell in May and go away if you like). The ability of Trump to deliver tax cuts and/or fiscal stimulus has to be question given his erratic behaviour.

Anecdotally my impression is that aging investors are more exposed to equities than historically or than prudent risk management would dictate, even in this interest rate environment, and this is a contributing factor behind current sunny valuations. Any serious or sudden wobble in equity markets may be magnified by a stampede of such investors trying to protect their savings and the mammoth gains of the 8 year old bull market. For the moment through, to misquote Chuck Price, as long as the music is playing investors are still dancing.

Warring Wireless

The M&A permutations being talked about in the US telecom sector are fascinating. The AT&T/Time Warner deal has commentators frothing at the mouth about possible tie-ups, particularly given the laissez faire attitude to regulation of the new FCC Chairman, Ajit Pai. Names such as DISH, Comcast, Disney, Netflix, SiriusXM, Charter are regularly tied to the big telcos and/or each other in the speculation.

Existing wireless revenues from AT&T and Verizon have plateaued and are now starting to decrease due to cut throat competition on unlimited plans from T-Mobile and Sprint, as the graph below shows.

click to enlarge

The T-Mobile/Sprint merger is now been talked about again, after the year long FCC spectrum moratorium on competitors talking has passed. The current bidding frenzy between AT&T and Verizon over Straight Path’s millimetre wave frequencies that can be used to carry large amounts of data over short distances has brought the hype over 5G services to a new high. Verizon’s purchase of XO last year enabled them to lease XO’s 102 LMDS licenses in the 28 GHz and 39 GHz bands so it is somewhat surprising to see them be so aggressive in bidding for Straight Path.

The analyst Craig Moffett believes that the winning bidder for Straight Path will have significant leverage with the FCC in determining how these wave frequencies are repackaged for use by competitors. Although wireless margins are not under significant pressure, as the graph below shows, it is obvious that there is now a full scale war for control of the wavelengths that will be critical to 5G services in the search for new revenues. T-Mobile and DISH are also holders of high frequency wavelengths been touted as suitable for 5G.

click to enlarge

The fascinating thing about the frenzy over Straight Path is that the hype over high frequencies such as the 39 GHz band is not new. The reason such spectrum has gone largely unused in the past is that it has been historically difficult to create reliable, secure, mobile connections in those bands. Variable weather conditions, often something as common as a rainy day, can impact coverage in high frequencies. I wonder what’s changed?

ILS illuminations

Insurance linked securities (ILS) are now well established in the insurance industry. ILS as an asset class offer, according to its many fans, the benefits of diversification and low correlation to other asset classes whilst offering a stable and attractive risk/reward return. The impact of the new capital on the traditional market has been profound and wide ranging (and a much posted upon topic in this blog – here, here & here for example).

ILS fund managers maintained an “aggressive posture” on price at the recent April renewals according to Willis Re as ILS capacity continues to demonstrate its cost of capital advantage. ILS fund managers are also looking to diversify, moving beyond pure short tail risks and looking at new previously uninsured or underinsured exposures, as well as looking to move their capital along the value-chain by sourcing primary risk more directly and in bulk.

An industry stalwart, John Kavanagh of Willis Re, commented that “with results on many diversifying non-catastrophe classes now marginal, there is greater pressure on reinsurers to address the pricing in these classes” and that “many reinsurers remain prepared to let their top line revenue growth stall and are opting to return excess capital to their shareholders”. The softening reinsurance market cycle is now in its fifth year and S&P estimates that “even assuming continued favourable prior-year reserve releases and benign natural catastrophe losses, we anticipate that reinsurers will barely cover their cost of capital over the next two years”.

Rather than fight the new capital on price, some traditional (re)insurers are, according to Brandan Holmes of Moody’s, “deploying third-party capital in their own capital structures in an effort to lower their blended cost of capital” and are deriving, according to Aon Benfield, “significant benefits from their ability to leverage alternative capital”. One can only fight cheap capital for so long, at some stage you just arbitrage against it (sound familiar!).

A.M. Best recently stated that “more collateralised reinsurance programs covering nonpeak exposures are ceded to the capital markets”. The precipitous growth in the private transacted collateralised reinsurance subsector can be seen in the graph from Aon below.

click to enlarge

Nick Frankland of Guy Carpenter commented that “the capital landscape is ever-changing” and that “such capital diversity also elevates the position of the broker”. Some argue that the all-powerful role of the dominant brokers is exacerbating market softness. These brokers would counterargue that they are simply fulfilling their role in an efficient market, matching buyers and sellers. As Frankland puts it, brokers are “in the strongest position to provide access to all forms of capital and so secure the more beneficial rates and terms and conditions”. Dominic Christian of Aon Benfield commented last year that “to some extent alternative sources of capital are already, and have already uberized insurance and reinsurance, by bringing increased sources of supply”.

Perhaps alone amongst industry participants, Weston Hicks of Alleghany, has questioned the golden goose of cheap ILS capital stating that “some new business models that separate the underwriting decision from the capital provider/risk bearer are, in our view, problematic because of a misalignment of incentives”. Such concerns are batted aside as old fashioned in this new world of endless possibilities. Frighteningly, John Seo of ILS fund manager Fermat Capital, suggests that “for every dollar of money that you see in the market right now, I think there is roughly 10 dollars on the sidelines waiting to come in if the market hardens”.

As an indicator of current ILS pricing, the historical market spread over expected losses in the public CAT bond market can be seen in the exhibit below with data sourced from Lane Financial. It is interesting to note that the average expected loss is increasing indicating CAT bonds are moving down the risk towers towards more working layer coverages.

click to enlarge

In a previous post, I argued that returns from an ILS fund index, with the net returns judgementally adjusted to get to comparable figures gross of management fees, were diverging against those from a pure CAT bond index. I argued that this divergence may illustrate that the ILS funds with exposure to the private collateralised reinsurance sector may be taking on higher risk exposures to pump returns (or may be passing risks amongst themselves in an embryonic spiral) and that ILS investors should be careful they understand the detail behind the risk profiles of the ILS funds they invest in.

Well, the final 2016 figures, as per the graph below, show that the returns in my analysis have in fact converged rather than diverged. On the face of it, this rubbishes my argument and I have to take that criticism on. Stubbornly, I could counter-argue that the ILS data used in the comparison may not reflect the returns of ILS funds with large exposure to collateralised reinsurance deals. Absent actual catastrophic events testing the range of current fund models, better data sources are needed to argue the point further.

click to enlarge

In their annual review of 2016, Lane Financial have an interesting piece on reducing transparency across both the public and private ILS sector. They characterise the private collateralised reinsurance sector as akin to a dark pool compared to the public CAT bond market which they likened to a lit exchange. Decreased transparency across the ILS sector “should send up warning flags” for all market participants as it makes calculating Net Asset Valuations (NAV) with monthly or quarterly frequency more difficult. They argue that the increased use of a relatively smaller public CAT bond market for pricing points across the ILS sector, the less credible is the overall valuation. This is another way of expressing my concern that the collateralised reinsurance market could be destabilising as it is hidden (and unregulated).

In the past, as per this post, I have questioned how the fully funded ILS market can claim to have a lower cost of capital against rated reinsurers who only have to hold capital against a percentage of their exposed limit, akin to fractional banking (see this post for more on that topic). The response is always down to the uncorrelated nature of ILS to other asset classes and therefore its attraction to investors such as pension funds who can apply a low cost of capital to the investment due to its uncorrelated and diversifying portfolio benefits. Market sponsors of ILS often use graphs such as the one below from the latest Swiss Re report to extoll the benefits of the asset class.

click to enlarge

A similar exhibit, this time from a Lombard Odier brochure, from 2016 shows ILS in an even more favourable light!

click to enlarge

As anybody who has looked through any fund marketing metrics knows, performance comparisons with other investment strategies are fraught with bias and generally postdictive. The period over which the comparison is made and the exact indices chosen (look at the differing equity indices used in the comparisons above) can make material differences. Also, the size and liquidity of a market is important, a point which may negate any reliance on ILS returns prior to 2007 for example.

I thought an interesting exercise would be to compare actual historical ILS returns, as represented by the Swiss Re Global Cat Bond Total Return Index, against total returns (i.e. share price annual change plus dividends paid in year) from equity investment in reinsurers across different time periods. The most applicable business model for comparison would be pure property catastrophe reinsurers but unfortunately there are not many of them left.

I have chosen RenRe (RNR) and Validus (VR), from 2007, as representatives of the pure property cat business model, although both have diversified their portfolios away from pure short tail business in recent years. I also selected three of the biggest European reinsurers – Munch Re, Swiss Re and Hannover Re – all of which are large diverse composite reinsurers. Finally, I constructed a US$ portfolio using equal shares of each of the five firms mentioned above (RenRe represents 40% of the portfolio until 2007 when Validus went public) with the € and CHF shares converted at each year end into dollars.

The construction of any such portfolio is postdictive and likely suffers from multiple biases. Selecting successful firms like RenRe and Validus could validly be criticised under survival bias. To counter such criticism, I would point out that the inclusion of the European reinsurers is a considerable historic drag on returns given their diverse composite footprint (and associated correlation to the market) and the exclusion of any specialist CAT firm that has been bought out in recent years, generally at a good premium, also drags down returns.

The comparison over the past 15 years, see graph below, shows that Munich and Swiss struggle to get over their losses from 2002 and 2003 and during the financial crisis. Hannover is the clear winner amongst the Europeans. The strong performance of Hannover, RenRe and Validus mean that the US$ portfolio matches the CAT bond performance after the first 10 years, albeit on a more volatile basis, before moving ahead on a cumulative basis in the last 5 years. The 15-year cumulative return is 217% for the CAT bond index and 377% return for the US$ equity portfolio.

click to enlarge

The comparison over the past 10 years, see graph below, is intriguing. Except for Validus, the CAT bond index beats all other firms and the US$ portfolio for non-volatile returns hands down in the first 5 years. Hannover, Validus and the portfolio each make a strong comeback in the most recent 5 years. The 10-year cumulative return is 125% for the CAT bond index and 189% return for the US$ equity portfolio.

click to enlarge

The comparison over the past 5 years, see graph below, shows that all firms and the portfolio handily beats the CAT bond index. Due to an absence in large loss activity over the recent past and much more shareholder friendly actions by all reinsurers, the equity returns have been steady and non-volatile. The 5-year cumulative return is 46% for the CAT bond index and 122% return for the US$ equity portfolio.

click to enlarge

Overall then, ILS may offer less volatile and uncorrelated returns but I would personally prefer the, often lumpier, historical equity returns from a selected portfolio of top reinsurers in my pension pot (we all could have both in our pension funds!). Then again, the influx of new capital into ILS has put the future viability of the traditional reinsurance business models into question so future equity returns from the sector may not be too rosy.

At the end of the day, the bottom line is whether current market risk premia is adequate, irrespective of being supplied by ILS fund managers or traditional reinsurers. Based upon what I see, I have grave misgivings about current market pricing and therefore have no financial exposure, ILS or equity or otherwise, to the market at present.

Additional Comment, 29th April 2017: The ILS website Artemis.bm had an interesting piece on comments from Torsten Jeworrek of Munich Re during their March conference call. The applicable comment is as follows:

“And now I give you another example, which is not innovation per se or not digitalization, but you know that more and more alternative capital came into the insurance industry over the last years; hedge funds, pension refunds, participating particularly in the cat business and as a trend that not all of the limits they provide, cat limits are fully collateralized anymore. That means there are 10 scenarios; hurricane, earthquake, [indiscernible], and so on; which are put together, but not 10 times the limit is collateralized, let’s say only 4 times, 5 times.

That means these hedge funds and pension funds so to speak in the future if they don’t have to provide full 100% collateralized for all the limits they provide, they need a certain credit risk for the buyer. The more they entertain, the more there’s a likelihood that this reinsurance can also fail. The question is how far will that go and this kind of not fully collateralized reinsurance, will that be then accepted as a reinsurance by the regulator or will that be penalized at a certain time otherwise we don’t have level playing field anymore, which means the traditional reinsurer who was strongly monitored and regulated and also reported as really expensive and a burden for our industry and for us and on the other hand, you have very lean pension and hedge funds who even don’t have to provide the same amount of capital for the same risk.”

Productivity Therapy

The IMF has sponsored another paper from staffers on the global productivity slowdown, with the catchy title “Gone with the Headwinds”. The paper reiterates many of the arguments concerning advanced economies referenced in this post, such as total factor productivity (TFP) hysteresis due to the boom-bust financial cycle and resulting capital misallocation, “an adverse feedback loop of weak aggregate demand, investment, and capital-embodied technological change”, elevated economic and policy uncertainty.

click to enlarge

Also cited are structural headwinds including a waning information and communication technology (ICT) boom, an aging workforce, slower human capital accumulation, and slowing global trade integration (including the maturing of China’s integration into world trade). An exhibit on the ICT trends from the report is reproduced below.

click to enlarge

The report highlights short term remedies such as boosting private sector demand, efficient spending on infrastructure, strengthening balance sheets, and reducing economic policy uncertainty. Longer term remedies cited include policies to boost technological progress, policies to mitigate the effects of aging, policies to encourage migration, advancing an open global trade system, exploiting policy synergies, structural reforms, raising the quantity and quality of human capital.

Now, how many of these remedies are likely to be pursued in the current populist political environment? Although Trump has shown signs recently of doing the opposite to what he fought the election on, overall it does look like we are merrily going down a policy dead-end for the next few years in important advanced economies. Hopefully the policy dead-end will be principally confined to the US and they wouldn’t take too long in figuring out the silliness of the current journey and the need to get back to trying to deal with the big issues intelligently. Then again….

Crimping CDS

The post-crisis CDS market has undergone significant regulatory change including a substantial regulatory overhaul due to the Volcker Rule, requirements from reporting to central clearing under the Dodd–Frank Act and the European Markets Infrastructure Regulation (EMIR), and Basel III capital and liquidity regulations. Measuring the size of the market consistently is notorious difficult given different accounting treatments, netting protocols, collateral requirements, and legal enforceability standards. Many organisations have been publishing data on the market (my source is the BIS for this post) but consistency has been an issue. Although a deeply flawed metric (due to some of the reasons just highlighted and then some), the graph below on the nominal size of the CDS market (which updates this post) illustrates the point on recent trends.

click to enlarge

The gross market value (defined by BIS as the sum of the absolute values of all open contracts with either positive or negative replacement values) and the net market value (which includes counterparty netting) are better metrics and indicate the real CDS exposure is a small fraction of the nominal market size, as per the graph below.

click to enlarge

Critics of the regulatory impact on the liquidity of the CDS market argue that these instruments are a vital tool in the credit markets for hedging positions, allowing investors to efficiently express investment positions and facilitating price discovery. A major issue for liquidity in the market is the capital constraints imposed by regulators which impedes the ability of financial institutions to engage in market-making. The withdrawal of Deutsche Bank from the CDS market was seen as a major blow despite some asset managers and hedge funds stepping up to the mark.

The impact of rising interest rates in the coming years on the credit markets will likely have some interesting, and potentially unforeseen, consequences. With a plethora of Goldman Sachs alumni currently working on Trump’s “very major hair cut on Dodd-Frank”, amongst other regulations, it will be interesting to see if any amendments lead to a shot in the arm for the CDS market. Jamie Dimon, in his most recent shareholder letter, calls for an approach by Trumps’ lieutenants “to open up the rulebook in the light of day and rework the rules and regulations that don’t work well or are unnecessary”.