Quantitative Methods – Page 21

Industry Ranks February 2012

Posted on 4 February 2012 by David Merkel

I?m working on my quarterly reshaping ? where I choose new companies to enter my portfolio.? The first part of this is industry analysis.

My main industry model is illustrated in the graphic.? Green industries are cold.? Red industries are hot.? If you like to play momentum, look at the red zone, and ask the question, ?Where are trends under-discounted??? Price momentum tends to persist, but look for areas where it might be even better in the near term.

If you are a value player, look at the green zone, and ask where trends are over-discounted.? Yes, things are bad, but are they all that bad?? Perhaps the is room for mean reversion.

My candidates from both categories are in the column labeled ?Dig through.?

If you use any of this, choose what you use off of your own trading style.? If you trade frequently, stay in the red zone.? Trading infrequently, play in the green zone ? don?t look for momentum, look for mean reversion.

Whatever you do, be consistent in your methods regarding momentum/mean-reversion, and only change methods if your current method is working well.

Huh?? Why change if things are working well?? I?m not saying to change if things are working well.? I?m saying don?t change if things are working badly.? Price momentum and mean-reversion are cyclical, and we tend to make changes at the worst possible moments, just before the pattern changes.? Maximum pain drives changes for most people, which is why average investors don?t make much money.

Maximum pleasure when things are going right leaves investors fat, dumb, and happy ? no one thinks of changing then.? This is why a disciplined approach that forces changes on a portfolio is useful, as I do 3-4 times a year.? It forces me to be bloodless and sell stocks with less potential for those with more potential over the next 1-5 years.

I like some technology names here, some energy some healthcare-related names, P&C Insurance and Reinsurance, particularly those that are strongly capitalized.? I?m not concerned about the healthcare bill; necessary services will be delivered, and healthcare companies will get paid.

A word on banks and REITs: the credit cycle has not been repealed, and there are still issues unresolved from the last cycle ? I am not interested there even at present levels.? The modest unwind currently happening in the credit markets, if it expands, would imply significant issues for banks and their ?regulators.?

I?m looking for undervalued and stable industries.? I?m not saying that there is always a bull market out there, and I will find it for you.? But there are places that are relatively better, and I have done relatively well in finding them.

At present, I am trying to be defensive.? I don?t have a lot of faith in the market as a whole, so I am biased toward the green zone, looking for mean-reversion, rather than momentum persisting.? The red zone is pretty cyclical at present.? I will be very happy hanging out in dull stocks for a while.

P&C Insurers and Reinsurers Look Cheap

After the heavy disaster year of 2011, P&C insurers and reinsurers look cheap.? Many trade below tangible book, and at single-digit P/Es, which has always been a strong area for me, if the companies are well-capitalized, which they are.

I already own a spread of well-run, inexpensive P&C insurers & reinsurers.? Would I increase the overweight here?? Yes, I might, because I view the group as absolutely cheap; it could make me money even in a down market.? Now, I would do my series of analyses such that I would be happy with the reserving and the investing policies of each insurer, but after that, I would be willing to add to my holdings.

Do your own due diligence on this, because I am often wrong.? One more note, I am still not tempted by banks or real estate related stocks.? I am beginning to wonder when the right time to buy them as a sector is.? As for that, I am open to advice.

Implications

So, given that the Industry Rank categories above come from Value Line, I went to their stock screener, selected the industries, and asked for all of the companies that:

are in their top 5 (of 9) categories for balance sheet strength, and
their horribly overworked analysts think can return at least 15%/yr over the next 3-5 years.

This combines safety, growth potential, valuation, and in my view, how promising industry prospects are.? Here are the results:

ABC ADM ADTN AKAM ALL AMAT AMX AOL APOL ARB ARRS BIDU BRKR BX CAH CBEY CECO CELL CKP CL CNQ CPB CPSI CREE CTRP DNR DRIV DV EBAY EDU EFX ERIC ESI FST GMCR GOOG HCC HRC IN INFA INTC ISIL ITRI IVC JNPR K KKR KR LIFE LRCX LTRE MASI MCHP MDCI MKC NFLX NIHD NILE NOK NTRI NVDA NXY ONNN OTEX QGEN QLGC QSII RAX RIMM RMD SHEN SOHU STM STRA SWKS SWY SYY T THG TMO TNDM TRH TRI TSM TSRA TUP TXN UNTD UPL UTHR VOD VOLC VZ WBMD WBSN YHOO ZBRA

When I do my next portfolio reshaping for clients in the next week or so, these stocks (and a few others) will compete against the 35 existing portfolio names for the 34-36 slots in the portfolio.

Full disclosure: Long HCC, INTC, THG, VOD

We Eat Dollar Weighted Returns — III

Posted on 2 February 2012 by David Merkel

Somebody notify the Bogleheads, they will like this one, or at least Jack will.? Yo, Jack, I met you over 15 years ago at a Philadelphia Financial Analysts Society meeting.

How bad are individual investors? at investing?? Bad, very bad.? But what if we limit it to a passive vehicle like the Grandaddy of all ETFs, the S&P 500 Spider [SPY]?? Should be better, right?

I remember a study done by Morningstar, where the difference between Time and Dollar-weighted returns was 3%/year on the S&P 500 open end fund for Vanguard, 1996-2006.

But here’s the result for the S&P 500 Spider, January 1993- September 2011.? Time-weighted return: 7.09%/year.? Dollar-weighted: 0.01%/yr.? Gap: 7%/yr+

Why so much worse than the open-end fund?? Easy.? Unlike the professional managers at Vanguard, and the relatively long term investors they attract, the retail short term traders of SPY trade badly; they arrive late, and leave late on average.

There is far more analysis to be done here, but to me, this confirms that Jack Bogle was right, and ETFs would be a net harm to retail investors.? The freedom to trade harms average investors, and maybe a lot of professionals as well.? It may also indicate that short-term trading as practiced by technicians may underperform in aggregate.? Not sure about that, but the conclusion is tempting.

One thing I will say: I am certain that profitable trading is not easy.? If you are tempted to trade for a living, the answer is probably don’t.

Anyway, here’s my spreadsheet on the topic:

Full disclosure: I have a few clients short SPY, hedged against my long positions.

On Junk Bonds

Posted on 27 January 2012 by David Merkel

If someone were to ask me my opinion on Junk Bonds at present, fool that he would be to ask me because I know real experts elsewhere, I would say this: They are good for a speculative trade, but dumb money has arrived.? Be ready to sell when the momentum fails.

High yield ETFs sell at decent premiums which leads to the creation of more units.? High yield closed-end funds — 73% trade at a premium.? You could issue a new high yield CEF, and come out at a lower premium than the current average.? I think I smell smoke.

Hmm….? If I owned junk bonds I would hold, and wait for momentum failure.? Buying now seems risky to me.? Most of the risk stems from global conditions.? We don’t know what will happen in the Eurozone. The rest of the risk stems from speculation.

I am a fan of junk bonds when nobody likes them, but there are too many fans now, and for bad reasons, most of which boil down to “I am old and I need income.? The fed has eliminated good choices for income, but I need income anyway, so get me yield.”

I had a conversation with a friend of mine in her upper 70s today where she asked “why are you suggesting I sell my funds that provide the most income?”? I said that I did not trust junk bonds at present and would look to lighten up, besides, the fund she owned has underperformed over the last 10 years.? If she really wanted income from junk bonds, I would look for a new fund for her.? So I am looking for a new HY fund, with an arm twisted behind my back.? It’s not the right idea, but she won’t listen.? (She’s not paying me.? I help my friends as best I can.)

The illusion of yield drives many older investors; they need income, and the delusional Fed thinks that low yields will yield prosperity.? It may make some people take more risk, but it will not yield prosperity.? There will be a lot of impoverished old people at the end of this, and they will be angry — at themselves, their advisors,? and the powers that be.

-==-=-=–=-==–=-=-=-=-=-=-=-==–=-=-=-=-=-=-=-=

This is not to say say that junk spreads are low; they are moderate to high at present.? But the spread relationship is manipulated by the Fed at present, making spreads seem high.? No market is truly free, but the Treasury market is affected by the Fed to a high degree.? The high quality bond market follows Treasuries closely.? Junk bonds don’t.? Junk bonds follow a hybrid of what Treasuries and common stocks are doing.? With stocks doing well, junk bonds run as well.

But we are still in an environment where more things can go wrong than right.? Until the US government figures out how to finance itself, we are in dangerous territory.? Given present political conditions, I don’t see how that works out; everything looks like a stalemate at present.

So be wary, and don’t overcommit to risk assets.? I would be neutral on risk assets at presemt, but ready to be bearish if there are problems in Europe or China.

Against Simple Valuation Metrics

Posted on 24 January 2012 by David Merkel

There have been a lot of articles dealing with use of corporate free cash flow lately:

Dividends — get them, are they sustainable?
Buybacks — do they add value or not?
Acquisitions — are they overpaying?? What are the synergies?

But you never hear about the last one — internal investment for organic growth.? There is a simple reason why — it is silent as night.? No one makes announcements on it.? If done properly, it is as quiet as a plant growing.

Dividends are simple — is there enough free capital to issue them, and do the other three priorities?? It is useful to ask how much room there is to increase the dividend, and how well the company can grow its earnings at the present rate.? Companies that pay a dividend understand that equity deserves a return, and are more careful with their capital as a result.? They often grow faster than companies that do not pay dividends.

But I never analyze a company primarily on its dividend yield.? I would rather look at the full set of the drivers of value.

Buybacks are harder because we don’t really know what the company is worth, and buybacks add value when you buy below the value of the company, and lose value when you buy above it.? In the reinsurance industry, it is understood that buybacks above 1.3x tangible book destroys value.? The threshold will be different in other industries because the value of intangibles will differ — but for industries where intangibles mean little, that 1.3x tangible book can be a useful limit.

We can do pro-forma analyses on acquisitions to see if they add value or not.? The best simple proxy is how large the acquisition is relative to the acquirer.? Small acquisitions typically add value? because they add a complementary product, a new marketing channel or region, lower costs, or raise product quality.

Large acquisitions typically lose value because acquirers overpay and integration is difficult.? One exception: negotiated sales by large private sellers.? There is no auction, and no winner’s curse.

The best acquisitions are small, but lead to an increase in organic growth.? Also, the best acquisitions are early; the worst acquisitions are imitative and late.? Typically the best deals get done first.

But much as I like managements who think that the equity deserves a return, via dividends and intelligent buybacks, the hard stuff gets done in organic growth: how are last year’s profits being increased on the existing infrastructure?? In mature industries, this is tough, which is why they typically return free cash flow to shareholders.? But when you find a company that can eke out improvements in a mature industry, finding changes that no one else does, hang onto that company, because it is driving profitable change in the industry.? (And probably taking share from others…)

The less mature the industry, the more room for organic improvement, and thus more free cash flow is dedicated to internal investment, and less to rewarding current shareholders.? In such a situation, it pays less to look at dividend yields, and more at dividend growth, adjusted for ability of growth to be sustained.

-=-=-=- begin rant mode -=-=-=-

This is why I am not crazy about simple articles that say:

Here are the five highest yielding companies of this industry, or
Here are the seven highest yielding investments of [famous investor, or company], or
Here are the companies that are buying back stock rapidly, or
Look at the combined dividend plus buyback yield of these companies…

Everyone wants to squish value investing into one simple metric and from what I have seen, it does not squish well.? That is one reason why I try to view companies off of the competitive dynamics of the industry in question, and adjust the metrics accordingly.? After all, no matter how cheap a company looks in an industry that is obsolete, like newspapers, it is rarely a good idea to buy.

Thus, I am skeptical of the many articles that are spit out by inexperienced investors that have a computer and can crank out a few simple ratios, and spew out some canned facts about a company — these articles are widespread, and not limited to writers on Seeking Alpha, or Zacks, or those that submit to Yahoo! Finance, and they have some canned and wrong way of identifying competitors.

Avoid these articles, and instead, look for some degree of qualitative reasoning — some depth that shows genuine industry knowledge, and not an ability to automate the provision of web “content.”

-=-=-=- end rant mode -=-=-=-

Maybe I should be quiet.? After all, the provision of bad advice on the web is a good thing for me.? The more people are misled, the better value investors with broader skill sets do.

But that’s not why I started writing on investments.? I was not a professional investor until I turned 39.? I read widely, and spent a lot of time reading the works of many different investors as I worked to develop a theory that encompassed most of it.? No, I don’t see how to encompass all of it… and what I can encompass is understood with some amount of error.

My view as I write is not so much to give “buy this” or “sell this” ideas so much as to get people to think differently about investing.? I recently looked at the amount of business/economics/finance/investment books that I have read over the past 25 (post-academic) years, and it would fill 3-4 bookcases.

So try to think of the companies that you own, or might own, like businesses.? Look at the dividends, and to buybacks at bargain prices, and analyze sustainability and growth prospects, but also look at opportunities for growth.? Many aspects of value can’t be encapsulated in simple ratios or rankings, but sadly, the majority of articles touting stocks will do just that, and for the most part, they are useless.

There.? I said it.? But it needs to be said.? The practical question to me is whether I should stop submitting my content to sites like Seeking Alpha, which to me have become a lot of noise, and which I wish I could get Yahoo! Finance to allow users to filter out of the news stream.

I let almost anyone republish my content, so dropping anyone would be unusual for me.? Or, should I drop all external users of my content, and allow no republishing?? If you have a strong opinion, submit it in the comments.? I’ve been a nice guy with all of this, but if you have good reasons for exclusivity, let me know, and I will consider it.

But to close I will say, look at a full range of valuation and performance metrics when buying a stock, and consider the industry dynamics to understand what matters most given the maturity of the industry.? That takes some work, but guess what?? Working intelligently and hard leads to better profits in investing.

The Rules, Part XXIX

Posted on 20 January 201220 January 2012 by David Merkel

Risk premiums should never be capitalized, they should only be taken into income as earned.

This may end up being another odd post of mine.? I’m going to start writing about bank regulation, but I will end up talking about monetary policy.

There are many people who hate the rating agencies. They hate them because they are a convenient target, and most people don’t understand what they do. Rating agencies provide opinions. Nothing more, nothing less.

Many people would like to get rid of the rating agencies. But it’s not that easy. Regulators outsource their credit rating function to the rating agencies because they don’t want to do that work.

There is a way to eliminate the rating agencies, and I have written about that before. But the idea is so radical, that the banks would rather have the rating agencies exist, than use my idea.

So what’s my idea? Simple. If you were setting up a portfolio, what would you assume would be the minimum that you could earn on the portfolio? My minimum would be buying Treasury bonds and earning interest on them.

So if I am looking at a portfolio of risky assets, I would split each asset into two. I would mirror the cash flow pattern of each asset, and construct an equivalent Treasury portfolio to mimic the cash flows. All of the cash flows above that amount from the risky asset are the risky cash flows. The amount of capital that banks hold as reserve against losses should be proportionate to the present value of risky cash flows.

Unlike my last piece on this, I am not saying that the whole present value of risky cash flows should be held as capital against losses. But the regulators should use this, if we are not using rating agencies, as a proxy for credit risk in bank asset portfolios.

Why is this a good measure of credit risk inside banks? The market for lending is fairly efficient. Debts that have more risk have higher interest rates.

This measure of risk benefits from the concept of simplicity. It can be applied everywhere. And, there is good theoretical justification for it. Any return that is upon the government bonds is subject to question.

But suppose we decided to use this as a major portion of our formula for regulating bank capital. What would happen to monetary policy?

Well, if the Fed tries to do something similar to ?operation twist” it would require banks to hold more capital against their positions, because the safe interest rate falls, it causes the risky portion of each loan to rise. As such, any sort of ?operation twist” would fail, because the rise in capital levels, would blunt any advantage from over Treasury interest rates.

From my vantage point, it would be a real plus to have monetary policy neutered in that way. The Fed, should it deserve to exist, should be concerned with the banking system and its solvency. It should not be concerned with the overall level of interest rates. If lowering interest rates lowers the judgment of solvency, then that would restrain the Fed from being too aggressive in lowering rates. And that would be good. The Fed has generally not succeeded with monetary policy. They have been too loose in the past, leading to the problems of the present.

And, as I have said before, we should not have unelected bureaucrats driving our economy, rather, we should have Congress do it because we can vote them out.

That’s all for now. Thanks for reading me. I appreciate all of my readers.

On Predicting the Future, Redux

Posted on 20 January 201220 January 2012 by David Merkel

From a reader, ptuomov:

If you run a regression of the magenta line on variables that have similar trends, you will get a spuriously high R2. I think you should try to explain the weekly changes in the magenta series instead. (I may have misunderstood you regression, in which case please show the actual data series in the regression so I?ll understand it better.)

Um, that’s not always true.? I did not get a Ph. D., but I passed my Ph. D. field in econometrics, including passing the oral exam.? I try to be really careful with regressions, unlike most.? I avoid multiple passes over the data, and I avoid “specification searches,” which are glorified hunts for correlations.

As it is, the regressors that I used are not highly correlated with each other.? They don’t have similar trends.? Here is the correlation matrix:

The regressors were very different variables, and were independently useful for deciphering the relationship.? Had it been otherwise, the t-coefficients would have weak, with the F-coefficient strong.? As it was, the t-coefficients were all strong.

This is not spurious.

Industry Ranks January 2012

Posted on 7 January 20127 January 2012 by David Merkel

I?m working on my quarterly reshaping ? where I choose new companies to enter my portfolio.? The first part of this is industry analysis.

If you are a value player, look at the green zone, and ask where trends are over-discounted.? Yes, things are bad, but are they all that bad?? Perhaps the is room for mean reversion.

My candidates from both categories are in the column labeled ?Dig through.?

Whatever you do, be consistent in your methods regarding momentum/mean-reversion, and only change methods if your current method is working well.

P&C Insurers and Reinsurers Look Cheap

Permanent Asset Allocation

Posted on 6 January 2012 by David Merkel

	Short run		Intermediate		Long Run
	Nominal	Real	Nominal	Real	Nominal	Real
Stocks	+	–	+ small	– big	+	0
Bonds	–	0	0	0	+	0
Cash	+	–	+	–	+	–
Gold	0	–	+ small	–	+	–




	Short run		Intermediate		Long Run
	Inflation	Real	Inflation	Real	Inflation	Real
Stocks	+	0	– small	– big	+	+
Bonds	–	–	0	0	+	+
Cash	+	0	+	0	+	0
Gold	0	–	+ small	– small	+	0

(Note: Nominal = Real + Inflation)

This article is meant to tie up some loose ends, and suggest the outline of what might be a clever way to do asset allocation.? Who knows?? At the end, there might be a surprise.

I’ve done two articles recently on the effects of inflation expectations and real interest rates on two asset classes in the short run — gold and stocks.? Tonight, I want to extend that two directions, to bonds and cash, and whether the effects aren’t different in the long run.

First, bonds in the short run.? Interest rates rise, bond prices fall.? Interest rates fall, bond prices rise.? Doesn’t matter whether that comes from real rates rising, or inflation.? That’s pretty simple, because most bonds are mostly interest-rate driven.

Second, cash in the short run.? Leaving aside financial repression, for the most part cash assets return in line with inflation.? Cash is simple… so what happens in the short run is also what happens in the long run.

Okay, now let’s lengthen the time horizon.? In the long run, gold keeps pace with inflation, nothing more, nothing less.? Bond returns rise if interest rates rise over the long term because of higher reinvestment rates for cash flow, and again, it doesn’t matter whether that comes from inflation or real rates.? Opposite if interest rates fall.

Think of 1979-82: by the time bond yields were nearing their peak levels, bond managers were making money in nominal terms with rates rising because the income from the coupons was so high, and it set up the tremendous rally in bonds that would last for ~30 years or so.

In that same era, stock multiples collapsed.? But eventually stock prices stopped going down even with competition from bond yields, because the earnings yields were so large that book values roared ahead, supporting prices.? That also set up the tremendous rally in stocks that would last for 18 years, until it finally overshot, giving us the present lost decade-plus.

But high rates, whether from inflation or real rates, presage high future bond and equity returns.

One nonlinearity here: in the intermediate-term, rises in real rates kill stocks, but rises in inflation nick stocks.? Why?? Inflation may improve nominal revenues at the same time that it raises the cost of capital, but rises in real rates indicate capital scarcity, raising the cost of capital with no increase in revenues.

=-=–=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-==-=–=-=-=-=-=-=-=-=-==-=-=–==-=-=-

Harry Browne proposed a “permanent portfolio” back in 1981, composed of equal portions of cash, bonds, gold, and stocks.? Reading about the idea in Barron’s in the late 1980s, I did not think much of the idea.? I think differently now.? After my last few articles on related issues, mentioned above, I realize that each of the four asset classes react differently to macroeconomic stimuli in the short run, with a lot of overshooting.? A mean-reverting strategy has a lot of power in this context, and it is double-barreled, in that it lowers volatility and raises returns.

My clients will receive the full details on this as an asset allocation strategy, but my readers have enough from this that if you want to do a little work you can figure this all out yourselves.

All that said, I am surprised at how well the strategy works.? Too easy, and easy strategies rarely work.

Stock Prices versus Implied Inflation

Posted on 5 January 2012 by David Merkel

Eddy Elfenbein wrote a good post recently on the stock market versus inflation expectations.? When I read it, I said to myself, “Wait, is the relationship between nominal and real rates really 1:1, or is it more complex?”? Though it is not certain, the regressions that I ran indicated that 1:1 was not falsified by the data.? The regression:

Inflation expectations determined the much of the value of the S&P 500 for the last nine years.

And you can see the relationship here as well:

The short answer is “yes, inflation expectations have driven stock valuations for the last nine years.”

I’ve been spending time on issues like this for a variety of reasons, and I’ll try to explain them in the near term, but that’s all for now.

On Insurance Stock Indexes

Posted on 3 January 20123 January 2012 by David Merkel

I’m still toying with the idea of starting an insurance-only hedge fund.? I own a lot of insurers, and I think that I get the better of that market.

Where I have a harder time is with what to short. Shorting is tactical not structural, and I am less good at the tactical vs structural.? Having a tradable benchmark to short against would be useful, but what exists there?

There is one ETF focused on insurance that has any significant volume — KIE.? In the past, it was capitalization-weighted, but now it is equal-weighted.? That stems from a change in the index that the ETF follows, from one set by KBW to one set by S&P.

Personally, I don’t get the change, but here are my statistics on the change:

The “Old KBW” column comes from segmentation done by KBW.? The other columns are done by me.? There are some matters for judgment:

Do you include Berkshire Hathaway?? I think you should.? Do you include foreign life insurers traded on US exchanges?? I think you should.

I am also more willing to place a company in the “Conglomerate” category because of companies that are in multiple lines of insurance, without a dominant area of insurance that they are in, or, they have significant non-insurance ventures.

Anyway, the new KIE overstates the insurers in Bermuda and the Brokers.? It understates life insurers and conglomerates.

Aside from that, the new S&P index, being equal-weighted, is more mid-cap than a whole market index would be.? Also, if I put more effort into this, I would segment companies into their proportions, and there we be no conglomerates.

These may be trivial concerns to some, but if you are thinking of running a portfolio that might be shorting KIE against other insurance longs, it makes a considerable difference.