Valuetrap13.files.wordpress.com



Energy Equities Are Cheap? Define That, PleaseDale WettlauferJanuary 4, 2016Last week, I saw someone say energy is cheap and if a company just survives this downturn, like banks in 2009, it’s going to be all good. I couldn’t disagree more, based on a few tenets:Energy is a classic cyclical industry best bought for long-term gains when it’s hated, there has been a substantial washout, and it doesn’t look at all cheap.Conversely, one of the worst times to buy energy equities is when the sector looks cheap. 10x earnings and 1x book on an asset base that wastes away at 10% to 20% per year with returns on incremental capital of 5% is not a cheap asset.Counting barrels, and putting values on developed and undeveloped reserves, is a relative heuristics game I’ve rarely seen married to actual cash economics of the exploration & production (E&P) industry.Like any equity an E&P’s value is the NPV of all the future free cash flows that can be pulled out of the asset. Counting barrels is the same proximate but misleading exercise as attaching EV / eyeballs on .com companies in the 1990s.Like any almost equity, an E&P can be modeled using the three primary financial statements and key fundamental drivers. I will offer three models at various pricing assumptions to demonstrate the sensitivity of an E&P’s value.I believe there are three overriding factors in valuing an E&P: Price of the hydrocarbon it produces, production costs, and finding and development costs, all expressed per barrel of oil equivalent (BOE). Price per BOE is largely out of the company’s control. It’s a commodity with different grades, but the same grade in the same play is going to be priced pretty homogenously (I’m admittedly simplifying here).Let’s look at real prices over the last 150 years and nearly 50 years:Exhibit 1: Real Oil Price, 1865-2015Source: British Petroleum, Charlotte Lane Capital (CLC)There are a few salients in this time period that should at least prompt an investor to consider how badly one can be mistaken in assuming an equity extracting this commodity would be a bargain just because the prices of the equity or the commodity are off by 25%, 50%, or 75%. If the commodity stays down for 10 years or more, you’re fighting a losing battle in digging out excess returns.Exhibit 2: Real Oil Price, 1968-2015Source: British Petroleum, CLC Energy equity outperformance off a peak isn’t an iron law; ExxonMobil beat the market for the 5, 10, 15, 20, 25, 30, and 35-year periods from the top in oil in 1980. So far since the 2008 and 2011 peaks, performance has been very poor XOM and E&Ps (which some might argue makes them coiled springs – God bless).A very large portion of XOM’s outperformance from 1980 forward was due to its dividend payout as well as, in my estimation, its more hedged stance as an integrated producer and refiner of energy products. Looking at crude vs. XOM from 1980 forward, a widow and orphans fund holding these shares with no idea of what was going on in energy markets might not have even noticed nominal and real oil prices crashed by 65-75% in the ensuing 5-6 years.Exhibit 3: XOM vs. Nominal Crude, 1980-2015Source: EIA, CLCHow about a stock with a different dividend policy and a far less hedged exposure to energy markets? Here it is:Exhibit 4: SLB vs. Nominal Crude, 1980-2015Source: EIA, CLCUnlike XOM, Schlumberger underperformed the broad equity market in every period (5-35 years in five-year increments and badly for stretches) from 1980 forward. While integrated energy services are a favorite short for Charlotte Lane in 2016 within an overall directionally-neutral sector strategy, I’m getting off on a tangential point here in explaining it’s risky to assume just because energy prices are down, you can produce excess returns in this sector by just buying and waiting. In a low inflation environment with a surfeit of reserve growth vs. consumption growth in the last ten years, that could get you killed, especially if non-OECD consumption growth stalls out.Many will point to production vs. consumption growth and say the market’s balanced…just you wait! Looking at it through that lens, sure, these two things move in tandem over time:Exhibit 5: Crude Consumption vs. Production, YoYSource: BP, CLCI don’t think this tells you much of what you need to know in terms of timing energy equities, though. Where’s the buy signal in the 1980s? In the beginning of the aughts? The sell 4-5 years ago? I believe a better indicator lies in looking at reserve growth vs. consumption growth:Exhibit 6: Crude Proved Reserves vs. Consumption, YoYSource: BP, CLCThis tells much more fully the classic commodity cycle story of overinvestment, consumption catching up to that overinvestment, and overinvestment then turning into underinvestment. Anyone who has covered a full cycle or two in commodities will recognize that (I would argue here putting the semiconductor analyst on commodities equities when the cycle turns up in 2020-2023 would be a good idea because that person will have iterated on numerous cycles in their career).Why are reserves so important? Can’t the owners of these reserves just sit on these reserves and wait for better prices? First of all, it’s not a forestry operation, where seedlings that grow into trees can just sit around and wait for higher prices. The capital devoted to harvesting lumber comes at the back end, close to the time of ultimate consumption. A barrel of oil takes $20-25 per barrel to find & develop (F&D), as of 2014:Exhibit 7: F&D Costs Per BOE, US Oily E&PsSource: Bloomberg, CLCAbout 20-25% of that goes into the exploration component and the rest goes into the development component. Once the resource is developed, that’s a ton of capital in the ground, given production per barrel of developed reserves runs at 10% to 20% annually. The overwhelming incentive once that cost is sunk is to get out of the ground all the cash possible, not sit on it unless you truly have a very, very long-term time horizon.We can think about this via a simple ROA decomposition. ROA = asset turns * net income. A dollar of assets that turns into revenue of $0.10 to $0.20 annually is very slow-moving. Costco this is not. A producer needs a 50% net margin in this scenario, not even counting corporate overhead assets, to produce an ROA of 5% to 10%. Lever that 2:1 and you get ROE of 10% to 20%. So what happens when oil prices decline? The only thing left to do is shut in the assets, wait, or speed up asset velocity. Major oil producers can’t elect the first two and evidently, national oil companies participating in cartels have a tough time electing those first two strategies as well. So the incentive is to drive returns via increased asset velocity. The spigots are open, in other words.This hasn’t been just a 2014-2015 phenomenon, either. While I believe Saudi Arabia is keeping up its output to deny cash flow to the Shi’ite regimes that surround it and of which it is deathly afraid, quite apart from my geopolitical ruminations, I think the microeconomic reasons this is happening are quite observable.Below, we see ROIC vs. WACC spread for oily E&Ps, which peaked in 2008, after which returns on incremental capital started to plummet. Not coincidentally, oil prices started to stall out and then imploded:Exhibit 8: Oily E&P ROIC vs. CrudeSource: Bloomberg, CLCForgetting all the Kremlin watching and trying to outdo Ian Bremmer on geopolitical matters, a fundamental investor just needs to consider it is a bad, bad scene in any industry when input costs (oil services, land, labor) and supply go up and price fails to follow suit. It all seemed so easy from 1999 forward, right? Why?One of the first things you learn covering a cyclical industry is there are leads and lags. Supply doesn’t respond to price signals immediately, whether because capital allocators are skeptical about the sustainability or due to physical or regulatory constraints governing the speed with which supply can respond. With China engaging in economic stimulus and a mercantilist expansion in the late 1990s forward, in response to slowing growth throughout the 1990s, and then Chairman Greenspan hitting the panic button very hard post .com bust, real demand and plus the cheap dollar just kicked off a huge global feeding frenzy in commodities.Every E&P, miner, equipment supplier, and services firm bought in (except XOM, heh, until XTO) once the ball was rolling and that’s where bubbles come from. Looking at just US E&Ps with $1 trillion in currently enterprise value, here’s the capex history and outlook (this is a constant group of companies that existed in each year measured in the chart below):Exhibit 9: Select US E&P Capex, 2000-2016ESource: Bloomberg, CLCNow, of course in a depletion-based business, the industry is going to have massive recurring capex, so it’s important to consider net capex (capex in excess of DD&A). This shows the net capacity additions for this group:Exhibit 10: Select US E&P Net Capex, 2000-2014Source: Bloomberg, CLCIn a global energy industry dominated by national oil companies (NOCs), the financial and physical flows of which are often opaque or total mysteries, I’m hardly scratching the surface on the global picture. If BP’s annual industry compendium is at all reliable (there’s always guesswork whether we use IEA, EIA, or other sources, and yes, BP did “overestimate” their own reserves), here’s a more comprehensive picture of supply potential relative to consumption:Exhibit 11: Years’ Crude Reserves, GlobalSource: BP, CLCAs I mentioned above, I believe supply potential (reserves, and more specifically, developed reserves) gives us a better picture of supply-side price pressure while production vs. consumption flows don’t tell us much, if anything, for the immediate or intermediate future. Below we see the two series: YoY change in crude price vs. the difference in YoY changes in reserves and YoY changes in consumption, with reserve changes lagged three years behind changes in oil price:Exhibit 12: Reserve Growth (Years’ Supply) vs. Crude Price ChangesSource: BP, CLCBelow we data from the same set rearranged into crude price change vs. (YoY change in supply less YoY change in consumption):Exhibit 13: Reserve Growth (Years’ Supply) vs. Crude Price ChangesSource: BP, CLCI will hasten to note here I’m not saying there’s a correlation here. There’s not – it’s too messy, with too many leads and lags, with too many inputs and withdrawals to claim there’s a correlation with any sort of R-squared. I would note the longer the precursor of production, reserves, grows at a faster rate than consumption, the chances for a production glut will grow and vice-versa. Looking at exhibit 5 vs. exhibits 12 and 13 is a good illustration. Exhibit 5 has a high R-squared and it says nothing about the outlook for the industry while the later exhibits are uncorrelated but speak to industry fundamentals to me. That is especially the case in conjunction with exhibit 8 and the real oil price charts that kicked off this piece. As soon as returns on incremental capital peaked and then went negative, it was time to get out of the way, short the sector, or at least get underweight for a long-only fund.As I said in my L/O career, I don’t want energy to color my (our) firm’s future either by being massively wrong on an overweight at the wrong time or by being naked the sector during a multi-year run. In my experience, generalists who favor asset light, non-cyclical businesses get this sector and other heavier sectors wrong by staying out for the good part of the cycle, regretting it, and then thinking they can pick off something with a good reputation at 10x EPS after a period of bad sector performance. To me, this is an absolute recipe for disaster. You can’t value E&Ps and oil services firms on price/book or earnings multiples. I never agreed with Michael Goldstein at Empirical Research on this, HOLT, or any other backward-looking mean reversion-driven algorithm. You must have some fundamental view of oil prices, industry microeconomics, and global macro. At the very least, if you’re going to play as a generalist, you have to look at charts like exhibits 1 and 2 to gain a long-cycle perspective. If those didn’t give you pause in the last few years, I’m not sure what would.These eight pages of preamble are all hypothetical. They are a general framework for testing concepts from both a top-down and bottom-up perspective. I’ll underwrite the worst company in the world if I like the cyclical setting, it’s bombed out, embedded expectations more than reflect that, and it’s trading at “N/M” EPS of ($1.50). On the opposite end, I’ll short it when margins and the cycle are pinned and it’s trading at 8x EPS. Earnings and price/book, and certainly EBITDA multiples often provide a 100% false indication of what you should be doing with a cyclical stock. You read this rule of thumb early in your career or you hear it from a veteran and you might violate it 1-3 times before it actually makes sense. After that, you go to school on it in the next cycle. For some, the penny never drops or they’ll have bad luck, so they say “I’ll stick to asset light, wide-moat companies that are non-cyclical.”These market participants often style themselves as savvy value people, but after getting bitten once or twice, they foreswear this type of company. Why would one want to write off a huge group of companies, especially when many value investors believe they have a behavioral edge against the crowd? This is a perfect group in which the patient, cycle-aware value investor can find a repeatable edge because it so often lends itself to herding and breakdowns in belief diversity.It takes more than heuristics on reserves and as I think I’ve emphasized, it takes more than P/E and price/book. “How else are we to value an E&P?” one might ask. After price received, there are two variables that are most important: production costs per BOE and finding & development (F&D) costs per BOE. Production cost drives EBIT and F&D costs contribute to EBIT, growth, and most importantly, capital velocity and ROIC.Let’s look at revenue, production cost, and F&D costs per BOE for a group of oily E&P companies:Exhibit 14: Revenue, Production Cost, F&D Cost / BOESource: Bloomberg, CLCAs we see, there isn’t anywhere near the differentiation in revenue and production cost per BOE as there is in finding & development costs. Revenue per unit is decided by hydrocarbon mix. Because most oilfield services work is outsourced, there’s no real moat on operating costs either. There are unitary returns to scale past a pretty low level in this industry, from what I’ve seen of Western independent oil companies. Given neither margin nor asset turns maintains primacy in the calculation of ROA, I believe the organizations that can demonstrate greater effectiveness in outperforming on F&D costs are superior E&P companies.This is why Costco is a superior company. They can price lower, live on reedy margins, and turn out returns on capital well above their cost of capital. It’s close to the fifth circle of hell to compete with this company. Same in the oil patch. If I can spend $12 to create a barrel of revenue-generating developed reserves while you spend $24, I can either generate twice the FCF for every dollar of net income each of us generates if our net margins are identical or I can generate two times the amount of developed reserves as you given the same net income.So, on to the model. The “Drivers” tab is 600+ lines, which would normally be absurd and a signal to push into the “too hard” pile any such company. With E&P companies, however, they have 2-4 hydrocarbons (crude, natural gas, and natural gas liquids, mainly) they express in 2-5 plays with sub-play disclosures. For each of these, there are multiple fundamental lines expressed in the footnotes. It quickly gets to a huge tab, but all of these things have meaning, so I put them into the spreadsheet. I don’t model forward each of these, however, just the most meaningful drivers. Those are:PriceF&D costCapexDeveloped reservesFlow rate (production / developed reserves)Production cost per BOE is captured in the income statement. Otherwise, every other lever is found on the Drivers tab. Line 611 of the tab drives capex and line 624 drives F&D cost per BOE. As you alter these inputs, reserve changes flow through the model, as of course do cash flow, the balance sheet, and the income statement. If you believe the company’s decline rates will change or the type of plays they exploit will change, you can alter the flow rates starting in line 149. Within the model, there are cells to monitor cash balances as inputs change. These are found below the capex line in Drivers 614 and on Income 149 (I use dividends as the cash distribution driver in out-years).The one thing going on in my E&P models are circular calculations. I otherwise despise circulars, but there are so many self-referential things in this industry it’s impossible to do a trial balance for each and then copy-paste-special into the “live” line. I do that in the calculations of interest income and interest expense, Income 107 and below.So, to the output of the model currently. I believe US E&Ps are discounting crude well north of $70. For EOG, which is the best in the industry, it takes crude of anywhere from $70 to $80 to fulfill expectations embedded in the stock today. With crude at $50, I get runoff value of $12 per share. With a crude assumption of $80, I get a value of $86 per share. With a WTI assumption of $70, ROIC surpasses WACC by a bit and return on incremental capital goes to WACC within six years.I’ll let others play with the models to their hearts’ content, but some cautionary words. Oil services costs and crude prices are very much linked. We see below Schlumberger revenue per active rig globally:Exhibit 15: SLB Revenue / Rig, GloballySource: Bloomberg, Baker Hughes, CLCLooking at a constant group of US E&Ps from 2000 through 2014, we see the same:Exhibit 16: Production Cost per BOE vs. CrudeSource: Bloomberg, CLCAnytime you make a big change in pricing assumptions, it’s best to change F&D costs per BOE as well as all the operating lines per BOE. As a general rule, I almost always drive P&L operating lines by units, even if I have to use notional units. Cost absorption in anything cyclical goes non-linear, which modeling via percentage of revenues often fails to catchI’d be happy to schedule a conference call to work through the models if anyone would like.Finally, a few additional observations. Charlotte Lane is long E&P and coal via Cloud Peak 2024 credit. Why would I do that given I just said E&Ps discount $70-80 oil? I have been running at 14% gross and nearly neutral in energy and will continue to do so because I do not want directional risk in energy. I believe integrated oil services firms and our natural gas liquefaction shorts have as much or more downside than E&Ps with less than half the upside in an energy rebound. Looking at exhibit 16, there is a ton of downside in oil services pricing for the customers of those companies to close the loop on increasing their returns on capital, both via opex and capex per BOE. This they must and will do if energy prices do not rise near-term.I will repeat a theme from the Q3 letter: taking directional risk on energy prices appears a needless stance given the decline in prices, the firm’s overwhelming short stance in industrials, and geopolitical risks. Islamic State is not an army capable of set-piece battles, it is led by a small band of psychotics, and it has built itself recruiting disaffected individuals living on a few dollars a day in a thoroughly impoverished region. However insignificant this group is in numbers, it controls financial resources and land mass; the potential to morph its command-and-control strategy into a cell-based strategy that was more effective under Al-Qaeda remains a wild card.A well-financed cellular organization could hit a vital terminal in Saudi Arabia, sink ships in the Strait of Hormuz, or, in the most dire scenario, set off a dirty bomb or chemical / biological device in a populous center. Whether that center is a hardened one, like New York City or Washington, DC, or a softer target such as Des Moines, Hartford, Oxnard, or Flint, the repercussions in global energy markets would probably be swift and severe. I do not need to have a short position in energy given this set of risks and given the overall (25%) short position the firm plans to maintain in Q1 2016.I believe low energy prices heighten these risks. It is no coincidence 9/11 happened after a 60%+ decline in real crude price from 1980 through the end of 2000. This attack was very likely funded (and executed, in part) by interests most threatened by this fiscal shortfall. Therefore, Charlotte Lane remains neutral on energy price directionally while having a significant ~15%+ gross exposure. I believe there is alpha to be had in pairing significantly overvalued shorts against less-overvalued ones (assuming scenarios from run-off to $80 oil across the sector). I know that may sound strange, but energy is a strange sector!Below is a pattern I have seen across commodity-oriented sectors:Exhibit 17: Schlumberger vs. CrudeCrude Left Hand ScaleSource: Bloomberg, CLCFrom 1995 through the beginning of 2015, these two series were 92% correlated. They have since come apart and SLB is trading where it last was in 2012, when Brent crude was near $110. Schlumberger is earning less than it did in that year and valuation has expanded rapidly. I believe the equity is pricing in a lot of hope that crude will rebound. This is another unintended effect of quantitative easing – it becomes easier to carry options on future events via equities. I have seen this in a number of commodity-related sectors with some demonstrated capitulation in 2015 in the likes of longtime favorite short Caterpillar, which we remain short.With Schlumberger, I am reminded of CAT. Their management is far more sober and I respect this company deeply. We owned it at the long-only firm where I worked and at which I incubated Charlotte Lane and sold it right after the company issued very bullish long-term guidance at its analyst day in mid-2014. But I believe hoards of generalist PMs have crowded into this name believing it provides a safe haven in the energy world without understanding how exposed this company is to commodity prices.Below we see the relationship between EBIT margins and crude:Exhibit 18: Schlumberger EBIT Margin vs. CrudeSource: Company filings, Bloomberg, CLCThe last time oil was at its present price, EBIT margins ranged from 8-14% (8% using real oil price) while current consensus calls for EBIT margins of 16% and 18% in 2016 and 2017. Regressing crude vs. SLB’s share price (not valuation technique, but just as a check), SLB should be at $30 today:Exhibit 19: SLB vs. CrudeSource: Bloomberg, BP, CLCI believe consensus is missing the potential for International margins to decline meaningfully in the next few years unless oil prices recover to $80+. The international operation at SLB (70+% of EBIT) is a very high quality one and the company maintains a learning curve lead it has maintained for many decades. However, cash margins across the E&P universe are in dire straits and customers will doubtlessly bring pressure to SLB to cut its prices. Contracts are longer in duration in International markets for SLB, which is why we’ve seen North American margins decline while International margins have hardly budged:Exhibit 20: SLB Margins by Reporting SegmentSource: Company filings, CLCSLB has already downsized its workforce by 20,000, or 15% of its total, so incremental margins held up well in 2015 as international contribution profits were very good. I have no problem seeing SLB EPS decline to $1.00 in a flat energy environment, so potential downside here is substantial. ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download