Good Day - I am currently developing a DCF Model to determine FV, PV, terminal value, and enterprise value. It works well for companies with positive historic growth (assuming the inputs are correct). I have been looking at a company with negative cash flow and having a hard time modeling negative cash flow. It is a young company which is why I am looking at it.If people could provide some insight on methods to determine future value each year, terminal, and enterprise value.For now what I have done for future cash flow calcs is assumed a crazy growth rate and the following equationIf previous year cash flow is negativeFuture Value = Previous Year FCF - Previous Year FCF * Growth Ratein the above calculation, the previous year FCF is negative.IF previous year cash flow is positiveFuture Value = Previous Year FCF (1+Growth Rate)Thanks,Jared
Try Professor Damodaran's site:http://pages.stern.nyu.edu/~adamodar/New_Home_Page/spreadsh....In particular, higrowth.xls:http://www.stern.nyu.edu/~adamodar/pc/higrowth.xls
In my DCF valuations I model cash flows explicitly over a 10 year period. So I start with current revenue, and then estimate revenue growth rate, operating margins, tax rates, depreciation, capex needs, and working capital needs over that 10 year period. For a free cash flow negative company to be worth investing in it must turn free cash flow positive at some point. So now you have to make a case for what will make free cash flow positive. Higher revenue? Expanding margins? Reduced capex? You can plug those estimates into years 1-10 (or however many years you want to explicitly model) and build a model for how free cash flow goes from negative to positive, and then estimate how much the positive free cash flows are worth today by discounting them back at the discount rate of your choice.Mike
Jared,Das Moose has some good ideas suggestions.The burning questions for companies in the red is cash burn and cash generation. The bottom of the statement of cash flows is an important number to track but it has to be in context to the whole statement. In a mature company the Financing Activities section is often given little more than a glance unless we are trying to figure out payout ratios or how much they are spending on stock buybacks. In a company that is trying to grow its way to profitability that little chunk of the statement may be the make or brake of the company. One way to play with this is Cash from Ops - CapEx +or- net borrowings.This is the money they are generating with the machine. It is a number more closely related to enterprise value then market cap. It also the number I would use to generate some return on metrics like ROC. This may sound odd to put debt into the top of the equation but in the high growth phase of a company access to capital is important. It is their estimated future returns plus their current cash generation that lenders are investing in and their ability to convince investors to invest is a type of return needed for young companies. So I would track both traditional ROC = EBIT*(1-tax rate)/(BV debt + BV of equity)andCash from ops - CapEx +or- net borrowings / (BV debt + BV equity)One other metric to track is the change in interest coverage ratio; where ICR = EBIT/Interest Expense (from statement of earnings). The raw number for these early stage companies is often ugly so tracking the change qtr/qtr and yr/yr becomes valuable. A pattern should be discernible if the pattern is acceptable then deviation from that pattern is either a warning or the signal for an opportunity. Ultimately what we are trying to figure out is when the cash burn starts to slow because of the returns being generated. Valuing a company who's yr1 earnings = -100, yr2 earnings = -200, yr3 = 250 is really difficult. When they get to the point where we have the opposite pattern and we can see the Statement of Cash flows improving along the same trend lines then we can estimate in what year they will be cash flow 0 and the next year when they are cash flow positive.Make sensejack
I have been looking at a company with negative cash flow and having a hard time modeling negative cash flow. It is a young company which is why I am looking at it.These types of companies are harder to value than others because essentially EVERY input is a guess.Probably the two easiest things to determine are the terminal value...basically, 2-3% (number isn't really critical), and the competitive advantage period (CAP). The CAP is not too hard to imagine in some industries. For instance in the CPU business (Intel, AMD), each "generation" is roughly 5 years. So the CAP is about 5-7 years. In others (Walmart for instance) it's harder to figure...but looking at for instance the rise and fall of Sears or K-mart from a historical perspective might give you some insight. At this point I'm figuring that Walmart is getting pretty close to the end. Amazon had a first generation and now (with Kindles) appears to be on the leading edge of the second run.Then you've got to estimate growth rates. This is kind of tricky to do but for a young company, their publications with respect to their business model will probably yield a lot of clues as to reasonable numbers to plug in to produce the usual sigmoidal shape.Then last for companies in the "negative" (development), you've got to look at their cash burn rate both currently and expected as they come into existence as a company.In the end the problem with these is that essentially you have to construct a marketing estimate. But it's not impossible. For instance MCP (supposed to come online this year) will ultimately settle down to roughly a $30/share stock. Right now the speculators have driven it up to $50/share, and Chinese investors doing pump-and-dump on the market with their announcements about restricting exports of rare earths which is highly unlikely to happen given the Chinese economic situation (read John Mauldin for some idea about international trade trends though recognize that he's a permabear).So I used this to buy in the $30's, rode it up to almost $60 when China made another one of their market manipulation announcements, sold a protective put (because it's hard to predict these momentum investment things), and laughed all the way down to about $45 as I stayed locked with most of my profit intact.
Everybody, thanks for the posts. Since my original post I have looked over the spreadsheets recommend by Moose and have been trying to learn the higrowth DCF model. I have a lot to learn. I have a number of the inputs figured out, but have some questions below.1) Is the are good rule of them to estimate stable operating margin. I ploted the historic data from the income statement in an attempt to estimate what the operating margin might be in the future. I have been using 5-15 percent depending on the company. Anythoughts?2) The higrowth model asks "Do you want to use corrent working capital as a percent of revenues for the future?" (Yes/No). Not sure what assumption to use here in %. Any suggestions/reference information would be helpful.3) Another question in the sheet, "How ould you like capital expenditures to be estimate 1) Grow at same rate as revenues? 2) Lag by two years? 3) Based on a fixed sales/capital ratio.Not sure which option I should model since they make a huge difference in the valuation. If somebody could explain this to me I would appreciate it. I have tried entering in a ratio based on the following formula, (Sales to Working Capital = sales / working capital). As an example using the above formula I get a stock value of $56/share, vs if I choose (1), I get $-26/share. Thanks,Jared
Jared,I doubt anyone can give you a fixed answer for your questions. There is a reason you have options. The game is to try to build a model that you believe best fits the firm being analyzed. All these variables can differ from industry to industry and management team to management team. If you really want to figure this out you need to find some comparable companies and dig through their histories and see what has occurred in the past. Then you need to decide how likely the new kid on the block is going to mirror that/those patterns. For some start ups operating margin expands over time. Software companies are like this; it takes time to build the marketing infrastructure and sales force, to get their foot in the door they may sell their product as cheap as they can. As they gain market share and the back end build up subsides operating margin expands. If they gain pricing power they gain further margin expansions. A retail firm is a different beast as margins are expected to be tight and growth needs to come from the top line. Service firms are different still and lie somewhere in the middle, and on and on. Some industries are far more working capital intensive then others. Some firms may need large working cap as percentage of revenue early and be able to ease that percentage down over time others by the nature of their business they are going to have to keep their foot on that gas pedal.The cap ex question may be the toughest because it really has two components, the nature of the industry and managements strategic planning. Some capex must be spent in order for the business to run how much above that is the management team pumping in to manage growth?Not a lot of help I know. Baby cash flow negative firms are tough to value using hard core fundamental approaches because so many inputs are educated guesses. This is why most of the realm short cuts to relative valuation methods. jack
Hi Jared,Jack gave you some great answers. I'd just like to add that forecasting working capital, and especially capex are still difficult, but critical, even in stable companies. So you might want to practice trying to model a more mature firm in your desired industry. Not just to get a feel for what your start up might do, as Jack suggests, but to understand the effects changes in these assumptions can have even on companies that are simpler to forecast.Cap Ex has always been the crux of my DCF efforts. It's where I end up applying most of the logic, and it's the most prone to skewing the forecast. Mine is a work in progress...Handling capex properly in your model reveals how a decent company can still be a horrible stock, if the industry economics require that they continually pump cash back into the business. There are models that look mostly at Operating Cash Flow, but that's like looking at your salary, without factoring in required large expenditures like cars and houses.I know you didn't ask, but:One of my two favorite investing finance books is Investment Valuation, by Aswath Damodaran, the author of those spreadsheets. I recommend getting a copy.http://www.amazon.com/Damodarans-Investment-Valuation-2nd-se...He is a professor at Stern School of Business, and offers video of his classes free online. (My other favorite is also a valuation book: http://www.amazon.com/Valuation-Measuring-Managing-Companies... )-joe
Any time I'm trying to get a real solid handle on a line item on one of the three sheets I tend to do the following.1)Pull or input the quarterly data of that line item for at least one business cycle. Currently I would probably reach back to 2005 - 2006. 2)I leave the above data in place as one set, I then create a rolling trailing twelve set. (trailing qtrs 4,3,2,1 then 3,2,1,4 and so on)I can then scan by eyeball or build a chart that lets me see the two sets of data. What I want to know is what "normal" is, what are the seasonal patterns. Were their any outliers and if so what was the cause?To get inside the "head" of an industry you may have to do this exercise for several competitors. Honestly, for many mature industries there often is not a lot of use out of this exercise because their numbers vary about as much as glacier does through the season. The kick in the head is the only way to find out who is who is to jump in and start rooting around. I dodge the CapEx issue as a direct adjustable/forecastable input because I'm lazier than Joe is. If it looks good/normal I grind it through an easy FCFF cruncher and live with the output. To put it another way I don't start with the revenue line. But I'm a mark with chalk cut with an axe guy. A better handle of a the money machine can be built, if you understand the moving parts, by doing it the long way.In the case of a young company it may be wise to take a good look at the PPE line on the balance sheet and how the industry tends to depreciate assets. CapEx, depreciation and PPE are all related. Getting CapEx right while getting depreciation wrong is going mess with your FCF outputs. jack
Sometimes you click the button when you didn't want to.A better handle of a the money machine can be built, if you understand the moving parts, by doing it the long wayI just want to clarify that this more moving parts model may better model how money moves through the company this should not be confused with creating a more accurate output. All IV models are flawed. If we know where the weaknesses are then we can make some educated Kentucky windage adjustments for the flaws. jack
I dodge the CapEx issue as a direct adjustable/forecastable input because I'm lazier than Joe is. If it looks good/normal I grind it through an easy FCFF cruncher and live with the output. To put it another way I don't start with the revenue line. But I'm a mark with chalk cut with an axe guy. A better handle of a the money machine can be built, if you understand the moving parts, by doing it the long way.Lazier, or perhaps wiser. I mark with a (searching for something really tiny and accurate here.... um, a laser), but I also cut with an axe. And a blindfold.Meaning, I get much more value out of the forecasting process, understanding the moving parts, than I do from the final fair value output number. Most of the time I run something through my DCF spreadsheet, and by the time all the tweaks make sense, I don't even bother with the "perfect number". The biggest mistake you can make with a DCF is to believe the output. Best to run several forecast scenarios, and develop a range of fair value.Now Jack is probably just farther along the curve than I am. I still find value in doing it the hard way, but someday a lot of that might be second nature and I can apply shortcuts.
Now Jack is probably just farther along the curve than I am. I still find value in doing it the hard way,I doubt it. I too find far more value in the process than the final output which is why each company I own has a hand crafted, no cookie cutter, spreadsheet. What we focus in on probably differs by personality, experience and approach. My EVA output always changes when I get pickier about what goes into ROIC, IC and WACC.(stupid notes) Rather than mess with a fancier FCFF formula I create side metrics which track the bits and pieces that I think are pertinent. As an aside I appreciate Dr. D's argument that in the end EVA = FCFF = FCFE if properly adjusted but I like the divergence, it gives me greater insight. I'm am better at aiming and shooting quicker than I was years ago. Occasionally the hand crafted spreadsheet is built after initial purchase. CAT for example was a no brainer I had been waiting a decade for. Now with the more detailed spreadsheet I can keep watch on them so I can decide to hold through a downturn or sell out before we get there and pick them back up on the other side. Do I reinvest dividends or let them pile up in cash?jack
My EVA output always changes when I get pickier about what goes into ROIC, IC and WACC.Here's some "picky" stuff, posted when I was trying to figure all this out for the first time:Calculating ROIC part 1: Invested Capital:http://boards.fool.com/calculating-roic-part-1-invested-capi...Calculating ROIC part 2: NOPLAT:http://boards.fool.com/calculating-roic-part-1-noplat-279663...-joe
You work harder than I do.IC = Total Assets - C&E - short term investments - long term investments - other NICBL (Non Income Bearing Current Liabilities) -if I'm feeling frogy or the company really looks like it needs it I'll make an adjustment for goodwill-if they are a R&D heavy company I make adjustments for that.-if they are really cash and investment heavy I'll run them with and without long and maybe even short term investments <-- think MSFTFor NOPLAT I substitute NOPAT = EBIT*(1-tax rate).When I get down to customizing its more about tweaking IC than NOPAT although I am likely to run a rolling trailing twelve on NOPAT and dig through their "one time expense" history. Something along these lines is where I starthttp://www.valuebasedmanagement.net/methods_roic.htmlThe way I look at it is that A)this is a whole firm approach so both top and bottom of the line need to reflect that reality. B)Invested Capital is the capital at work in the business/industry they are in. C)Too many adjustments to earnings and one can get lost and confused. So IC = capital invested to make the thing work (capitalized leases can fit under NIBCL if one so chooses) NOPAT = EBIT scrutinized for goofiness *(1-tax rate or estimated/average tax rate)if EBIT has some goofyness it can be rolling block smoothed too goofy and it gets roundfiled. I have better things to doEVA = (ROIC - WACC)*ICMVA = sum of the future value of EVA + cash - debt (as a whole firm number it needs to be adjusted for equity value)jack
Give me a minute while I scoop my brains up off the floor. I think the take away here is to read the book by Dr. Damodaran. What are your thoughts on using a very complicated model like the higrowth model which appears to be fairly customizable vs. a more simple model that would eval. a few variables?Thanks,JAred
Jared,The most important thing you can take away from this sideways discussion is that the journey is more important than the model or its output. Every model I build is founded on that premise. I want to know what makes that company tick and building the model is an excuse to, parse their three sheets, dig through notes, read press clippings and read analysts reports. The goal of all that legwork is to build a model that reflects the company, has inputs and outputs that are meaningful in relation to that company. Its your choice, if you can understand a model with many moving parts and it helps you understand the important business aspects of the company being studied then build the complicated model. If we are going to study a biotech firm we need to recognize that R&D is the heart of what they do. There are many reasonable ways to treat R&D some more complex than others. If you can gain insight by treating R&D in a simple manner than that is as far as you need to go. If you feel the need to gain better understanding by using a more complex method then that is the direction that is best for you. If I was to approach a biotech, the science is mostly way out of my wheelhouse, I would use a pretty simple cash flow or EVA approach and build a set of metrics that helped me track R&D's impact to the bottom line(actually EBIT for me). Another way to do that would be to input some portion of that R&D into working capital or invested capital. What we are looking for is, in essence, return on R&D because that is what pure play biotechs do. R&D is important at Cummings (CMI) but it is not the heart of the company. A diesel engine is a diesel engine. We can squeeze a bit more out of it in fuel efficiency or power but the engineering physics is pretty much understood and the sideboards well known. Capitalizing and/or tracking R&D is pretty much a waste of time. If Cummings can squeeze 4hp more out of their motor Navistar and Cat can too so there is little lasting competitive advantage. Activision needs to develop new titles only a very few of which will become "franchises". How valuable is their library? Is that library of titles value reflected in their Goodwill or in some other asset category? COGS is really straight forward in any software company so it needs no special treatment or attention, does it look like most of the industry, good, these are not the droids you are looking for. SG&A is something to eyeball because if they have to spend more on sales we have issues in the development pipeline. The long and short is use the method that makes sense to you and that you can later make sense out of. Often I do not know what I'm going to build into my model until I've done a fair amount of leg work. Make sense?jack
The most important thing you can take away from this sideways discussion is that the journey is more important than the model or its output. Every model I build is founded on that premise. I want to know what makes that company tick and building the model is an excuse to, parse their three sheets, dig through notes, read press clippings and read analysts reports. The goal of all that legwork is to build a model that reflects the company, has inputs and outputs that are meaningful in relation to that company.Well said, Jack. I wholeheartedly agree.
It has been awhile since I have posted to this thread. I have been reading up on Damodaran and playing around with one of his models. I have gone through the balance sheet, cashflow, and income statement for a company, read through their annual report a bit (this part is a bit painful) and started playing with the model. I think I have started to settle on the inputs I have chosen. I have no idea if my conclusion is correct of the input assumptions are reasonable. Can anybody recommend a way to check these assumptions to confirm if they are reasonable?The company I decided to value is Sunpower. Currently I value them at about 4x's what they are trading. Makes me a bit nervous. Here is what I have done so far. Using a high growth model using FCFF.1. Forecasting Revenue based on growth estimate.2. Growth estimate is based on two calculations. 1) Assuming constant ROC and reinvestment and 2) My estimate of the market in 2015 and what share they will have.3. Capitlized R&D4. Brought operating leases into Debt.5. Estimate a Beta based on competetors and risk.6. Included equity options in the valuationAreas I am not sure about.1. Estimating Cap Ex in the future. Should I be using current WC as function of Revenues for future cap ex or should I provide an estimate. Not real comfortable with what this number should be or how I determine a reasonable one.2. Operating Margin. What I ended up doing was using an average operating margin for 6 companies. Again I don't know enough about companies to know what is reasonable3. Debt to Capital Ratio. Absolutely clueless what I should use for stable growth. The number has a large impact on the valuation. Currently using 36.3% based on my estimate for the current period.Any suggestions would be appreciated. I realize that a paragraph is not going to give enough information for people to comment on my approach. I mainly want to see what I can do to verify my assumptions.Thanks,Jared
Jared,A couple ways to double check inputs are to use Dr. D's industry averages. For example, he has a spreadsheet that will help derive CapEx using inputs but it also has industry averages. One of the reasons I stopped using modified versions of his spreadsheets is because some of the inputs seemed unnecessarily complicated. Too many moving parts for this mark with chalk cut with axe knucklehead. It is important to know why he has them in the spreadsheets so you can make an intelligent substitution or elimination. One way to test your model is to play with a few inputs until your output matches current prices. This reverse engineering of market price can give you insight into the market. With both sets of inputs in hand you can make a rational decision about which output is more mostly right than the other. jack
Jack - Thanks. I have thought about building my own model which would be more simple than his model. I think his model is complicated so that you can toggle a number of switches to determine model inputs. My thought was to use a model that worked to start to understand what information should be included in a model and what it should look like. Would you recommend building my own model? To check the model based on today would I just go back in time say 5 years a ago use the financial info from that year, default rates from that year etc.. or do you have a nother method? Regarding the industry averages would you suggest using those as my inputs or checking those averages against the cash flow ouput for 10 years?Jared
Hi Jared,I highly recommend building your own model. For me, the model is useful as a tool, but I learned much more creating it from scratch than I ever would have using Dam's sheets. My copy of Valuation is dog-eared, but I've never used Dam's spreadsheets, except occasionally to try and resolve something I don't understand about the calculations.1. Estimating Cap Ex in the future. Should I be using current WC as function of Revenues for future cap ex or should I provide an estimate. Not real comfortable with what this number should be or how I determine a reasonable one.2. Operating Margin. What I ended up doing was using an average operating margin for 6 companies. Again I don't know enough about companies to know what is reasonable3. Debt to Capital Ratio. Absolutely clueless what I should use for stable growth. The number has a large impact on the valuation. Currently using 36.3% based on my estimate for the current period.You've correctly identified the hard parts :)1. I use Dam's industry average Marginal-Sales-to-Capital as a guideline. Then I assume (pretend) that there is a direct correlation of revenue to capital required (growth ain't free). Say MS/C is 3.0, then if I forecast revenue growing from $200m to $240, then the cap-ex required to do so is (240-200)/3.0 = $13.33m.Along with Dam's average MS/C, I also calculate actual historical MS/C for this company, and actual historical revenue to beginning of period unadjusted invested capital. Unadjusted meaning no adjustments for operating leases, capitalized R&D, etc. I eyeball all these stats to decide on (swag) a reasonable MS/C for the forecast.2. Similar to #1. I eyeball Dam's industry average EBIT Margin, as well as actual historical EBIT Margin for this company. If you want to include specific peer data, that's good, just make sure they're comparable. Different industries' margins vary wildly.3. Might depend on the nature of the business. Compare to several mature peers. Try to get a sense of how management views debt - do they have any? Is it all revolving short term? Do they issue debt to buy back shares? Etc.My model allows for separate inputs for all variables for each year in the (10 year) forecast, as well as terminal. It can be useful to model debt moving steadily from current state to some stable level (your 36% isn't a bad guess, depending on the industry).I think I've said this before. Once I go to all the trouble of building a model from scratch, and then painstakingly developing a forecast, two things become apparent:1. I have a reasonable understanding of how business in general, and this company, work.2. I realize there are oh-so-many fiddly bits that greatly affect the estimate of value.Realizing that I'm apt to be very wrong on at least one of the critical guesses, and that if I could tell the future I'd be surfing the coast of my own private island ........ I put down the spreadsheet and pick up Jack's chalk and axe. And put on a blindfold.The process is absolutely critical, the results aren't. That said, the spreadsheet is also useful to track the company over time, as previously forecasted values become actuals.Really, you seem very much on the right track here, as you had to have answered a whole slew of questions to have ended up with these three remaining areas of doubt.Cheers,-joe
I have built my model and started valuating a couple stucks. Would anybody be willing to trade opinions / conclusions on a few stocks I have evaluated. Would like to see how my estimate compares to others.I have looked at the following.AMD: No buyTyson: BuyFord: BuyIf anybody has looked at these I would be curious to get more into the details. I am more than happy to discuss my approach.Jared
I've not followed any of these companies (oh Ford, wish I had known you better when trading for less than a cup of coffee).It might take a few days, but I'll give Tyson a whack. I used to follow Sanderson, a rival chicken producer. Post what you like, but I'm going to try to NOT read it until I've got some ideas of my own...-joe
Joe – Here is my biggest weakness. I don't know enough about company evaluation so don't have a good intuative sense of what makes a company a good investment. For Tyson I looked at whether the majority of analysts rated Tyson as a buy. They did so I decided to take a closer look.Here are some of my assumptions.Cost of Borrowing: 4.8% (Assumed default spread of 1.6%, added to T-Bond Rate)Beta: .8Market Risk Premium: 4%Growth Rate of EBIT = 3.5% per year and 3.5 after year 10Reinvestment Rate: Starting at 8% and changing to near industry average in 10 years (20%)For now I have excluded operating leases.It seemed to me that I have assumed a pretty conservative growth rate and normal reinvestment rate and they still appear undervalued, so a buy to me. Let me know if you need other information. Also feel ask questions. I am new to this.Thanks,Jared
Best Of |
Favorites & Replies |
Start a New Board |
My Fool |
BATS data provided in real-time. NYSE, NASDAQ and NYSEMKT data delayed 15 minutes.
Real-Time prices provided by BATS. Market data provided by Interactive Data.
Company fundamental data provided by Morningstar. Earnings Estimates, Analyst Ra