About now a good portion of you are hoping The Fly will shut me up soon for this prattle on fundies as clearly my view is at best wacked and distracting. Let’s do a turn on valuation to get folks really tired of me.

Just what is something worth? The enormity of this science immediately excludes me from its serious contemplation. You think you know something and suddenly someone else comes along with a bigger balance sh…sorry, an economic event occurs that turns your thinking upside down.

In normal times, the three most common ways to value a stock are DCF, SOTP and relative valuation. I will leave it to a bigger brain to yap about game theory.

DCF

This leads to a clear and obvious value that should be free of question by way of Discounting future Cash Flows, or valuing tomorrow’s cash earnings today. Sadly, an honest DCF almost always leads to a price much lower than the current price. Everything is clearly not a short (?!?), so a DCF valuation is usually tweaked to suit the price target of the analyst. A valuation built on a straight multiple is subject to the hopeful whims of the analyst, but a DCF gets the extra juice of a subjective discount rate. Google DCF for a more detailed description of what goes on here.

I find it humorous when an analyst pops off with a DCF valuation. The vast majority of sell-siders do not publish cash flow statements and only casually mention cash flow yield whenever the conversation suits them. This usually occurs about the time a company does something dilutive to stockholders or valuation gets out of hand and the analyst does not want to shift the rating. Almost none of them discuss the nuances of the cash flow statement, either because they have no clue or because it would make it very clear the company is burning cash. The multiple equity raises by EMKR and AMSC over the past decade would have been perfectly clear before they occurred if a cash flow statement had been maintained – maybe such a discussion would have jeopardized a position on “the team” when the time came to sell stock?

Oddly, a DCF valuation is rarely presented by itself. An analyst will usually pair it with a straight earnings multiple. Again, discretionary, but inline with comparables.

SOTP - Sum Of The Parts

A company with disparate parts is valued based on the parts’ supposed market value. Often this is done when one or more of the parts are at risk of being sold off. However, like with DCF, this can be tweaked one way or the other to suit the analyst’s target.

SOTP works as long as the correct discount rate is used. This discount rate is different from the one used in DCF. Here we look for a discount to SOTP to support the effect of taxes on a sale. A common midpoint for a discount rate that I have seen is 20%, meaning that the market value of the company is 20% lower than the SOTP. A higher discount rate supports a bearish position and a lower discount rate supports a bullish position. However, a flat discount rate of any kind on SOTP does not account for the likely different tax locales of the various businesses or if one of the businesses is valued on its earnings.

A prime example of the negligence and laziness demonstrated by a sell-side SOTP can be seen in the recent sale of BG’s fertilizer business to Vale. The herd failed to take the time to consider the tax rate that might be exacted on the sale. After the deal was announced but before completion, Morgan Stanley went out of the way point out how the price BG received was below market and a sign of BG’s distress. Because the tax rate was so low, the after-tax return to BG was far greater than any sell-sider imagined and likely higher than such a sale performed by any other company in the space. It was all a “surprise” because the sell-side would not give mgmt a call to discuss how the deal would work.

Relative Valuation

Consider the following comp table:

A comp table is a list of stocks grouped by a common industry so that an investor may look at the differences in various multiples of theoretically similar companies. The idea is to identify what average multiple is appropriate for group and individual stocks. A long/short strategy will look at a comp table to identify stocks that are too expensive or too cheap with respect to the group and pair one against the other. A longer term view will look at historical multiples for the stocks and the group to calibrate the current multiple and identify if/when a bubble is forming.

A comp table must be built with Consensus estimates. Putting aside my beef with the quality of the numbers within Consensus, the resulting average leads to very consistent multiples. Even better, if you know how Consensus works and are able to apply a smidge of common sense to forward estimates, you can develop a pretty accurate view of the direction a stock will take in the near term regardless of whether the stock is currently a darling or a pig.

Reiterating the point: this must be done with Consensus estimates. You will see scads of different comp tables published by the sell-side but almost all use internal estimates and not Consensus. There must be an explicit reference in the note that numbers come from Consensus or you ignore the values in the comp table.

Most comp tables held by the buyside are automated. They sit in an Excel file (for example) with Consensus estimates and company info such as debt, cash and equity populated by Factset or Thomson or some other data provider. I would bet some machines trade off these inputs, but I would be surprised if machines immediately trade changes during earnings season given the volatility exhibited as the various estimates are populated. Because comp tables are automated like this, there is a case to be made for trading off the automated table even when you know the numbers are wrong. After all, this is theoretically what everyone else is looking at, right? Madness.

Some stocks trade on EV/EBITDA instead of P/E due to a heavy debt component in the balance sheet. The entry of cash and debt info into the Factset or Thomson databases is even slower than it is for Consensus estimates. These entries have a history of error so Factset and Thomson slow down the entry of cash and debt info until quality control has a chance to verify the numbers are correct. Thus, if there is a huge change to debt or cash in a company trading off EV/EBITDA, the retail investor taking the time to note the effect on EV/EBITDA has an upper hand.

The relative valuation method has been the most reliable and consistent in my experience. There are as many different views as to why this works as there are investors, but the fact is that it works best at determining an expected multiple to rate a stock.

There are other ways to value stocks. The most recent fad is to tag a stock with a target that is cycle-peak EPS times the cycle-trough multiple. These fads always end when a company and industry continues to put up increasing earnings and the stock continues upward. The herd will change its direction on multiples when the market keeps going up, just like DORs mandated that their analysts take down multiples when the market was contracting in 2008 and 2009.

A fun characteristic of stocks is that they will go up as long as the company posts consistently higher earnings. If the multiple stays flat, the stock goes up on the strength of earnings. Also, the multiple tends to increase slightly with each successive earnings increase.

Our investing goal is to find a stock that we think will go up in price (or down if we insist on shorting). There are tons of different ways to accomplish this goal. When using fundamentals, the goal is to find a stock that you feel is too cheap and buy it on the expectation that it will post earnings to push its multiple back in line with its group in a form of mean reversion. Another approach is to buy a stock that is appropriately valued but will post increasing earnings that will push the stock higher. Both require an understanding of a company’s earnings capacity and its current valuation as viewed by the entire investment community.

Swinging for the fences, we want to identify a stock that is grossly undervalued and will post heroic earnings increases. We can use the above methods of valuation to determine which stocks are super cheap. Then we can use our Superman-like forward analysis skills to figure out when the company will beat estimates again (presuming the stock fell down for a reason).

An interesting way to determine cheapness is to do a little balance sheet analysis. The best description for one of the more successful methods has the horrible name “Magic Formula Investing” or “MFI.” (www.magicformulainvesting.com) Pick up the book if you have no experience here. It is an easy read and will give you the foundation to move forward identifying cheap stocks with good earnings potential.

The problem with balance sheet analysis is that it tends to look backwards. It is very rare that an analyst has a forward-looking balance sheet, and the general lack of a cash flow statement – let alone an honest forward-looking one – means that most balance sheet commentary from the sell-side is flawed.

Incorporating forward looking information is the best way to make balance sheet analysis work. Even better would be to add forward info that contains investor sentiment (…Consensus…). For example, take the latest list of stocks from the MFI website and throw the tickers into one column in an Excel spreadsheet and put their associated stock prices in the next column. Next, add a column for this year’s (FY1) Consensus EPS. If possible, add another column for next year’s (FY2) Consensus EPS. The next column should create the forward P/E for the stock. The last column should create PEG. Ranking the tickers by positive forward earnings, low PEG and low P/E whittles the MFI list down considerably to help focus research efforts.

Wow. Check out how JPM changed estimates on his downgrade yesterday. Let’s compare it to what I wrote about Consensus over the past week or so:

- Sell rating but FY10 estimates above Consensus…check

- Getting more negative by lowering near quarter estimates but raising later quarter estimates, keeping FY10 estimates unchanged overall…check

I am calling it a weak downgrade. I have not read the notes directly (I have not been able to see them yet), but my guess is there is no reference to a company visit that led to such changes in his estimates. I would love to hear directly from folks who have actually read the notes to be sure he did not mention a visit with mgmt.

(update - finally saw the notes, no discussion with mgmt. this is just a macro call.)

I have ultimate respect for RC and his call on FSLR. The chart looks horrible. His call keeps me from stepping in long just yet. But if the premise for downside is entirely based on this JPM report then the call is at risk. There is no buysider who is willing to bet against FSLR’s presumed low ball guidance for FY11 which leads to something between $7 and $8 – a long way from JPM’s newly introduced $5.58 estimate. The folks who will buy this stock on increasing earnings reports and closer proximity to FY11 will not come off this view until mgmt tells them otherwise. Assuming things don’t fall apart (it looks like even Blansett thinks things get better for FSLR in 2H10), we should not expect mgmt to back off FY11 guidance until December at the earliest.

Worse, if you are short FSLR on this report and see the stock moving higher afterward, you are a little worried. I don’t mean being short via the apr10 90. I mean being short at least $10mln stock in the $100-$105 range when an easy valuation on FY11 guidance gets you to a minimum of $105. A 20x multiple on EPS at the high end of guidance gets you to $160.

So for a buyside short at $105 to work, you need to think mgmt has lost control of its business or that we see another 4Q08.

I am not long yet. I would prefer to see how the stock performs closer to $100, but I am not entirely sure we make it down that far.

A long long time ago, in a universe far far away, Consensus estimates were free and available to all who desired via I/B/E/S (http://en.wikipedia.org/wiki/Institutional_Brokers’_Estimate_System).

I don’t mean just average sales and average EPS. I mean detailed estimates by analyst. Ahh, glory days.

What happened? Thomson bought I/B/E/S and it all fell behind a pay wall. Then they bought Reuters. Factset also provides Consensus info but has always charged for it.

At least as far as my laziness has been able to determine, detailed Consensus estimates are hard to find absent some sort of surrendered account info. I think I worked pretty hard at it for a week or so a couple of years ago. Since then I sort of meander around things that might suggest free estimates. All in hopes of gaining access to

• sales
• gross margin
• operating/ebit margin
• earnings before tax
• net earnings
• EPS

…in detail by analyst at no cost. Dammit, I am entitled.

Since I am not entitled, even if I live in France (which I don’t, as far as you know…), I must find Consensus info on my own. So I surrendered some account info.

As a cheap bastard, why do I care about the elements of Consensus? What exactly is Consensus?

Consensus is a straight average of available analyst estimates and nothing more. Analysts submit estimates via notes and Thomson and Factset load their databases accordingly. If an analyst’s note has sales, gm, ebit, net, and eps info, it is all added to the Consensus database. If an analyst note has only sales and ebit info, only sales and ebit are added to the Thomson engine from that analyst.

Now that Thomson owns I/B/E/S – not free anymore – it must solicit folks to add estimates to its database. And get this: Thomson quality control is such that if you submit an estimate that is way out of some statistical boundary (I have no idea what that boundary is), Thomson will exclude your estimate from the average. If you have access to Thomson, you can see the actual estimates excluded from the average. Yet we have already determined that folks who submit notes at the end of the quarter could possibly have legitimate estimates that are way off the statistical average and still be excluded from the Consensus average. Consensus can be a very flawed number with regard to near term potential reports.

(To be fair, part of the reason Thomson does this is to prevent the effect of transposition errors. Transposition errors still occur more often than they should but the effect is minimized.)

More important than the elements of Consensus is “What is it?” For shorter term fundamental investors, changes to Consensus form the basis of all stock price movement. (Those familiar with the effects of derivatives or short interest know this view is badly flawed.) For longer term fundamental investors, Consensus sets the general price level around which yield is developed. It does not matter whether you like the sell-siders who input to Consensus or if you disagree with the actual input – Consensus sets the foundation. Much like all the “investors” who bitched about last year’s fucked up market but were too afraid to buy or short, if you do fundamental analysis and do not provide input to Consensus you have no business whining about how fucked up the estimates are today.

Let me rephrase the last paragraph. As a buysider whose business depends upon fundamental analysis, you cannot ignore the foundation of Consensus or the fact that you have no input into that foundation. You are an observer making decisions upon what you see, not on things you desire or can change. The mere presence of your fundamental model that says CRM is overvalued at anything higher than $40 does not mean the stock is an immediate short. This sounds an awful lot like a tech analyst talking about charts. The difference is that fundamentals provide a major input to the shapes tech analysis calls out and, as it so happens, fundamentals have their own weighing machine called Consensus.

Whoa. Freaky shit there. If the physicist in you wants to go one step further to encompass the effects of derivatives and short selling into some grand unified theory of stock prices, you will eventually center on the idea that the emotional whim of the last balance sheet focused on a stock really determines its local price range. I encourage you to step away from that ledge, my friend, especially when you happen to hear some drunken buysider admit that his success is “due more to his balance sheet than really knowing where the fundamentals are going.” It will simply make you want to drink more and rah rah the Fed against HFT aggressors. Instead, take the blue pill and rest warm and fuzzy in the idea that Harvard and Wharton teach the best of the best that fundamental valuation is the end-all and be-all of stock prices.

But I digress. General uses and tricks of Consensus followed by the final justification of fundie analysis are much more interesting.

The process of updating Consensus usually takes a day after a note is published. If a company reports Monday afternoon and analyst reports with updated estimates show up Tuesday morning, Consensus is usually loaded by Tuesday afternoon or Wednesday morning. Thus, you can get a jump on the effect of a change to Consensus by tabbing all the estimate updates in the morning notes and doing the average yourself. Of course, you would not be the only one doing this, but you would have the ability to see exactly where Consensus is going before the market opens and join the initial fray. As Consensus loads that afternoon or the next morning, more and more people come on board with the update and further the trade. Each stock responds differently to this process so be careful not to impress the response of stock ABC on stock DEF without a little investigation first.

An interesting event to watch is when an analyst gets canned and the associated estimates are removed. If the analyst is way off on a limb, the removal of these estimates from the average can create a noticeable effect on overall Consensus.

Most seasoned sell-siders know how Consensus works and the effect on stock prices that accompany a change in Consensus estimates. You will notice that many old hats with sell ratings also have above-Consensus estimates, and vice versa for buy ratings. Seems a little backwards. The first reason is that if a company misses the above/beats the below Consensus estimates it strengthens the analyst’s case. If the company somehow does beat/miss, the update to Consensus from the analyst’s estimates will be minimized and help the analyst’s case by mitigating the overall change to Consensus. Finally, if the company does miss/beat, the analyst can lower/raise the estimates and pressure the stock in the direction of the rating.

Another interesting occurrence with Consensus is what happens when a company crushes estimates in one direction and the herd’s sentiment is placed opposite. If we wake up tomorrow after the YGE report and suddenly are presented with info suggesting YGE estimates should quadruple for the upcoming year, it is very unlikely that the herd will immediately raise estimates in an honest way supporting such a change. Call it embarrassment, negligence, whatever…an analyst will never make the full adjustment to estimates required from a surprise call. Unless something really horrible occurs you will rarely see more than a 25% change to overall Consensus after an earnings call.

For those who think fundie analysis is only for the privileged few, consider the “anchoring and adjustment” defense of tech analysis (brief description in http://en.wikipedia.org/wiki/Anchoring). This tells us that big deviations from a previous norm generally do not happen quickly. While this is no consolation to anyone who was daytrading long FSLR from 105 only to watch some balance sheet take out all market orders from 106 to 104 just before the sell order at 106.10, it does point directly to the effect mentioned above about Consensus estimates not moving as far as they should. As fundie investing encompasses a longer term/slower rate of change timeframe, “anchoring and adjustment” tells us that retail investors have a much better chance than the big boys of making big money on fundamental analysis.

It’s a fucked up world we live in.

Sell-siders are generally smart cookies who have an odd penchant for putting their names in lights. They live for a spot on CNBC or Bloomberg. They do a little jig for every press release in NYT or WSJ or whatever local business journal serves their stocks. A day without a phone call from a client sets off a wave of depression. If you tell one you never made it past page one of the recent note the analyst will question the purpose of life.

A sad consequence of this narcissism is they don’t know when to let go. They think flip-flopping like a fish on a stock call could lead to the view they really don’t know what the heck is going on at the company. They are correct. The end result is they are married to their picks until a really good exit presents itself.

Another big issue with sell-siders is they are not allowed to invest in their own coverage stocks. They generally don’t have time to use their own methods to trade stocks outside of their coverage and thus use ETFs or mutual funds to park money. Some are mandated by internal controls to only use mutual funds. Very obviously they don’t know the sting of losing money on a flawed thesis. When a call goes wrong the analyst writes a note, gets yelled at and goes home to sleep. When an analyst is consistently wrong, the analyst sometimes picks up more business because clients like the consistency of the wrong call (“Clegg upgraded CSUN? Time to sell.”). These are not people I want telling me how to invest my money.

A good sell-sider (as viewed by the analyst’s employer) is worth the client base and not the stocks or the picks. Do we really think Analyst A is going to screw Best Client #1 by reversing a call on MU, even if the company was dumb enough to dilute shares 15% for a dying business? I would not bet on that unless I knew the analyst had already gone marketing on the subject. A good analyst – one who is famous and paid well – is more likely tied to his client base and not a call to integrity.

I am being overly harsh. There are a few who are successful AND intellectually honest. Most sell-siders don’t make $2m/yr and most of these are hungry to attract a client base that will support that sort of pay. The up-and-coming analysts tend to have some pretty good info in their notes if you know how to find it. All in the name of attracting new clients. Poor saps.

All are required to publish new views or info in notes before they are allowed to talk about them with clients. As mentioned above, they don’t want to piss off clients by a quick reversal, so some will hide the change in a carefully worded note so as not to alert casual readers to the change.

So when do you listen to a sell-sider?

Let’s go back to last week’s discussion of earnings. Anything that alters the near term view for earnings is important. “Near term” is the stock’s current valuation time frame. There are two times you want to listen to a sell-sider and one additional trick to understand what is closer to fundie reality using sell-sider input.

1. When an analyst raises or lowers estimates AND changes the rating at the same time.

Check out a few of the articles in here, in particular the first one. They talk about the effect estimate and rating changes have on stock prices. The short of it? The biggest effect occurs if estimates decline AND the analyst drops the rating, or estimates increase AND the analyst increases the rating.

http://www.docstoc.com/docs/27564894/ARD-Oct-09

2. When an analyst quietly changes individual sales, margin or EPS estimates but does not necessarily change full year EPS estimates or the rating.

Look for big changes in estimates that aren’t necessarily called out by analysts. When the herd has estimates clustered about one number and some dude pops off with something out in left field, ask why. If it is Gordon Johnson at Hapaolim you just ignore it. Otherwise, read the note and see if the change comes after a visit with mgmt. If so this is likely a datapoint passed on by mgmt.

The craziest is when you see the near quarter estimate increased but the far quarter estimate decreased (or vice versa) in order to maintain the same full year EPS estimate. This is pure madness. It will bring tears to your eyes when you see how often it occurs.

Be wary of estimate changes that originate solely by macro analysis.

3. Ahead of an earnings call, look at the difference in analyst estimates between the post-call update and just before quiet period starts.

Check out http://www.docstoc.com/docs/27565144/Detail-Estimates for a list of analyst updates to Consensus EPS for NSM. Note the ~4c/5.3% increase to estimates for the folks who updated estimates since the last earnings call Dec11. (While I am not trying to say NSM will blow out Consensus on its upcoming report, this does suggest that NSM is not about to shit the bed.) This tells you a change has occurred and that the upcoming reported results are more likely to be near the recent estimate updates than those set right after the call. Sometimes recent estimates are ridiculously off the Consensus average. The vast majority of the time recent Consensus updates simply indicate the real direction of the upcoming report.

Ideally you spot a single, well respected analyst popping off with an oddball estimate after a company visit just ahead of quiet period. This is a gift that happens more often than fundie detractors think. Properly placed, such info can be a huge factor when playing quarterly calls.

This is all well and good as long as you have easy access to detailed estimates.

Despite the general panic which has marked this month’s entirety and the recent poor performance of my account, I’m pleased to regard February as a success.  I do this because, at present, the value of my property is just below what it was at the highs of January.  And, despite being there, I see that many of my holdings are off from where they were weeks ago.  Hedging and scalping several short and long trades has helped shore up my account.  All I need is for my few lagging positions (such as the price of silver) to regain its lost ground and I will find myself charging in to new, previously unchartered, lands of prosperity.

Last week I was irked when my broker, in an act of malfeasance, stopped me out of a large part of my BHI short for a small gain, just as (basically before) it plunged deeper into the red.  After calling them up to question, in my most unpleasant vehemence, what was the intent of such contemptible actions, I discovered the flawed inferiority of their short contracts, which leave room for the holders to walk away at practically any moment, unannounced.

This got me to thinking, however.  It is apparent the markets desire to go downward; yet, to suggest that we shall repeat the effects of the beginning of this recession is folly.  The beginning of the recession caught everyone off guard.  Now, however, people are on their toes.  If shrouded faces in the dark are withdrawing options from the table, particularly so that I can feel it, then I would mark that as the presence of intervention, gentlemen.

In short, this feels like a bear trap.  Feel free to test my theory; since I’m on to it, I’ll be sure to align myself on the side of the hunters, and be one of those firing rounds into the pit you convicted short types are even now being coerced into.

At any rate, my planning disturbed, once more, I covered the rest of my BHI willfully and all of my short NOV stake, for sizable profits.  I’ll update you on the quantity of the gain later, and whether it completely absolves my realized losses from the last few weeks, or if I still have some ground to cover.

Friends, Cain Hammond Thaler will not stand for mediocrity.  If you have followed me over the last year, you know my soul occupation is the agglomeration and perfection of business.  Sadly, the last month has led me to break even, or level off, if you prefer.  I will not permit this to continue.

The markets continue their nonsense dance across my face this morning, and I find myself longing for the yester years of 2008.  You asshole union types despised the radiance that was the 2008 stock market sell off, but I made so much wealth off the corpses of retirees in that span of time, I wouldn’t mind another run down.

Or what about the late months of 2009, and the gifts they brought?  I made a gargantuan sum of treasure off that run as well, watching my hoard of equities appreciate past the point of being fathomable.  I’d love for the markets to keep running higher from here as well.

No, all in all, the only thing I can’t stand is having every position I own kicking me in the sternum, breakfast, lunch and dinner, like adolescents acting out against the powers that be.  On the one hand, I’ve brushed off the last two weeks as if a 400 point run down was nothing but a tiresome nuisance.  On the other hand, I missed a great opportunity.  If this sort of happenings is how we are going to define the next few months of trading, I think I’d rather get black out drunk and skip them, thanks.

I hedged my portfolio (again) last week by adding NOV and BHI shorts, which are at present treading just under the surface.  Actually, it would be more accurate to say I shorted NOV and BHI.  I intend to make money off of these trades.  However, I kept the exposure on the light side.  I think I’ve got a feel for the moment and how things are behaving, but I’d rather feel left out than dread being caught in.

To the EU: go fuck yourselves.  I’ve spent the last decades despising you fucking socialist types.  At every corner my ears get swamped with your criticism on such a host of issues from our foreign policy, to our debt obligations and how we abuse those relationships, to how being American is the mark of waste and greed.

Now, we have Greece, Spain, and Portugal spiraling into oblivion, with Britain hot on their heels and only Germany seems to be in a position to abandon them all to a sticky end.  There’s irony there, if you care to look for it.

So let me ask you, Europe; where are your smart ass critics now?  If there’s one thing I can’t stand, it’s a hypocrite; so does it strike any of you that for all your incessant bitching about how unsustainable American capitalism and growth has purportedly been, the entirety of your countries were effectively built on overconsumption, feeling entitled to services beyond your abilities and shit ass decisions?

I’ve since stopped caring about a collapse of the euro.  Let’s pretend the euro does collapse and the dollar rallies, alright?  So what?  Such a development would help to ultimately secure the future of American companies.  The money lost would more than compensate for itself.  And, since an influx of foreign currency could also be expected to flee to treasuries, then perhaps our government would find sufficient financing to encourage the Fed to stop fiddling in our immediate affairs.

I’m willing to gamble that stocks would quickly pull out of any fall and shore back up to play sideways.

In fact, the only element of our country that would be devastated from a collapse of the euro is American labor.  I already hear more irrational complaining in a day about how China is stealing American jobs than I would ever care to resubmit here.  So do you fixed income sorts honestly think you’ll survive two?