The Dark Side of Valuation: India Business Forum

I asked for the wireless because I hate podiums I just can't stand them it makes me very very nervous but um first I looked at how much time I had and had 45 minutes and I'm so verose it takes me 45 minutes to get into any topic so I thought I'd pick something very specific I thought I'd talk about deferred taxes for 25 minutes if you think I'm serious you're sick I mean I there's no way I'm talking about deferred tax of 45 seconds but basically 45 minutes is not enough time to actually talk about any specific detail in valuation so here's what I'd like to do I'd like to talk about what I think makes valuations go bad I talk about the Dark Side of valuation I'm talking about bad valuations and let's face there are two things that are true about valuation first most people who claim to do valuation are not doing valuation they're doing pricing and let me be very specific about what I mean let's say you decide to buy an apartment or a house you hire a realtor right the realtor shows you the apartment then she gives you a number or he gives your number I don't be sexist about this and says this apartment is worth 10 million ask yourself a question how did the realtor come up with the number did he or she value the apartment or did he or she she priced the apartment the answer is here she priced the apartment right they looked at other apartments in the neighborhood they looked at what they sold for and said that's what the value of your apartment is saying that's real estate they're not sophisticated there pick up an equity research report what do you say you see a company you see a multiple you see comparables this company is cheaper expensive given how these other 15 companies are trading at and let's face it the realtor is actually on more solid ground talking about comparables than most equity research analysts are because there is no other company out there that's similar to the company you're being valued so you kind of throw up your hands and you tell a few stories and those kind trust me those 15 companies are just like your company so much of what passes for valuation out there is pricing so often when we talk about valuation challenges you're talking about pricing challenges and I understand your job is to price assets I'm not picking on you for doing that but don't tell me you're doing valuation but when people actually do valuation you think they do a pretty good job right all these books out there on valuation you've taken all these classes on valuation we have all these models on valuation all this data that feeds into valuation you think that the state-of-the-art valuation must be getting better but it's not you see some incredibly screwed up valuations and the question is why what what can we do to fix it so remember about a couple of years ago the uh head of m&a at a major Investment Bank in New York asked me to come and talk to his bankers and I said what what do you want me to talk they've heard enough from me already they said we'd like you to talk about valuation to them because they're doing some really bad stuff in valuation and I said it's pointless nothing I say is going to change because it's not that they don't know how to do valuation it's that they're in a process where screwed up valuations are going to feed out of the process if the process is screwed up the valuations are going to be screwed up as well so what I'd like to talk about today is the process EV valuation what I think of is the three demons that cause bad valuations and none of them have to do with metrics and models in fact I was thinking about um sexy title for this I I've discovered that sexy titles are the biggest part of selling books and selling presentations and I thought the Bermuda Triangle is a good way to describe this hey you heard of the Bermuda Triangle right ships would go there planes would fly in there and just disappear this is where valuations go to die and here are the three sides of the triangle there's the biased side the biggest enemy of good valuation is bias and I'll talk a little bit more about this the second is uncertainty as human beings we're terrible at dealing with uncertainty we prefer to deny that it exists push it behind the curtain what uncertainty I don't see it out there and the third is complexity complexity in the sense we have all this data and we can build all these big models and after a while we lose sight of what we're trying to do bias uncertainty and complexity so let me start with the first of the sides of those triangles which is bias we'd like to think that valuation is an exercise in analysis you're trying to Value the company and come up with the number I wish it were true because here's what I tend to see I tend to see people think of a number number first and then come up with the valuation to back up the number they already have it's backwards right it's it reminds me of Alice and Wonderland remember Alice wers in the Queen's court and the queen says verdict first trial afterwards which might not be such a bad idea given some of the trials we've seen but that seems to be the State ofthe art and valuation value comes first valuation to follow so that's the first first thing when we talk about buyers it's value first the second is you almost never sit down to value a company with a blank slate in other words you already have preconceptions about companies in fact my valuation class how many of you been have taken my valuation class okay remember you had to pick a company early in the semester he one semester I tried a little experiment the day people picked a company I said before you start our valuing the company why don't you tell me with the company name whether you think you're going to f mind this company to be undervalued overvalued they complained they said but we haven't done the valuation yet I said just give me a gut feeling and they were pretty good about so I got the spreadsheet together of companies they had picked and their priors what they thought they were going to find and then they turned in their projects at the end of the 15th week and I did a correlation between what they thought they were going to find and what they found and here's a [Music] miracle most of them found what they thought they would find okay it it's I think something we don't think through but it's there and here are some of that think the the ironies in valuation when you when you when you sit down a value company what are you told read up as much as you can about the company find out more about the company that's good right but there's a dark side to that the more you know about a company the more biases you have about the company some of my easiest valuations have been of companies I don't know know anything about in fact when I go into a market off and I pick it last last year or two years ago I had to go to Slovenia do a presentation and valuation I picked Slovenia because my wife's grandparents came from Slovenia I wanted to see the country realized it was a little cantonement you could drive through it in 20 minutes right so I had to pick a Slovenian company and I'm sure you probably are aware of major Slovenian companies but I knew of none so I actually went to Bloomberg I picked a list of Slovenian companies I closed my eyes I pointed to the screen and I picked a company company called kir never heard of it before I didn't even know what they did one of the cleanest valuations I've ever done cuz I had no I had no biases I could sit there with the numbers look at what they did and draw a judgment on the valuation about 8 months ago I put a post up on Apple and some of you might have read this but I made a confession I said I've never been able to value apple as a company cuz I'm too biased I love the company and I know I love the company and it's going to show it's going to find its way into the numbers the more you know about a company the more difficult it becomes for you to put distance between you and the company and if you get to know the managers of the company God help you you play golf with the CEO don't even show me your valuation I am not interested you've got to maintain enough distance to be objective and that's not easy to do so let's talk about the sources of bias okay so if you think about where this bias comes from some of it is behavioral right which is you don't even know where it comes from sometimes look at a company I really like this company why I like the name of the company okay so some of it you just don't even know where it comes from some of it you can trace back if everybody else is saying good things about a company it's kind of difficult for you to step back and be objective and say bad things about that company do a Hur Behavior mentality in fact let me use use my my my class again to illustrate a concept many people who take my class this is the first time they've done a full valuation of a company and when you value a company and you come up with the value what's the first thing you checked against the market price right and if your value is different from the market price what's your second Instinct especially if you've never done valuation before which is I must have done something wrong let me go revisit my valuation valuations magically move towards the market price and the more and the less confidence you have in this process the more likely it is that is to happen and it's actually it makes sense because if you think about it if you're going to screw up it's always good to have lots of company right and being at the market price basically means hey I screwed up but so did everybody else that's a her behavior the other is hindsight Biers I have never ever used a case in evaluation class it's pointless and here's why if you use a case it's said back in time right so I ask you to Value AOL for Time Warner in 1999 you know what you're going to find time War to pay too much for AOL surprise and you know why you find it because you know what happened after 99 but I tell you when I give you the case act like you're 1999 what universe is that going to be on it's impossible to take out hindsight bias from this process if you're going to do valuation it has to be with real companies in real time because once you know the outcome All Is Lost so that's the first source of the second is what I call the power of suggestion I I remember talking to a managing director about he he was in charge of an IPO team did valuations and I said no and he was complaining about by the fact that these IPO valuations didn't always seem to come back with with numbers that he knew already off the top he said why why am I using these guys always and I said what do you do and he said I usually give them the company and I give them a rough sense of what I think the company's worth and then I asked them to do the valuation so basically in the company and said this is an IPO I think it's worth about $15 per share but do the valuation anyway right and not surprising they finished the valuation and what comes back $15 per share you can it's very easy to throw in subtle ways of pushing numbers so when I give you a company and in fact I tried this experiment once I actually gave two groups I was teaching two valuation sessions must have been torturing myself that semester okay and I gave them the same company to value for an IP and in one class at the time that I gave them the numbers I gave them a little subtle hint I said well I think you know the $4 per share is is what people thought was worth but ignore that I said that and then I show them the distribution of valuations in each class and magically the $4 became the middle of the distribution the class where I threw it in but here's the biggest source of bias you tell me who pays you and how much you get paid for doing a valuation I'll tell you which direction the bias is and how much the bias is the first rules in valuation right who pays you basically deter which direction the bias is I give the story every time I teach my valuation class if you've heard the story before you're going to hear it one more time I tell them about a company called Lin cable this about 20 years ago and it was a time when AT&T had this option to buy it was a cable company obviously and there's option to buy 49% of Lin cable at an appraised value so that's where the option was written so the time for the option to be exercised comes about AT&T hires Morgan Stanley so you you guys are going to be Morgan Stanley right maybe you already work for Morgan Stanley but if you're not so and your job is to assess the value of the 49% so that AT&T can buy that for so you work for the buyer Lyn broadcasting hired Leman Brothers I'm sorry to do this to you but think of yourself as pre-b bankrupt Leman Brothers but your job is to assess the value of the same 49% so that L gets so you work for the seller you work for the buyer two Investment Banking teams go to work grind through the number numbers come up with models now they come back with two very different numbers one side comes back with $105 per share the other comes back with $155 per share now who do you think came back with $105 per share why because you work for the buyer you did your job you came into the low ball number you did your job you came into the high ball number in fact the numbers were so different they decided to call him a third investment banker why set for two fees when you can have three I guess and they call in was parella and I'm going to say something incredibly harsh about these guys but I mean every word of it these guys couldn't value a $20 bill in a brown paper bag if you put in front of them they' invent some multiple that nobody had seen before Enterprise Value to cash in the bag 3.3 before you know you'd be paying $66 for a $20 bill but if you was the steam par you're right in the middle right you don't want to piss off either side too much so 105 55 where's the safest place for you to be 130 right they came back with $127.50 I'm going to let you in on a little secret evaluation don't let it out of this room if you're ever asked to Value something never ever come back with a nice round number don't tell me the Target price is 40 tell me it's $38.8 it's amazing what that second decimal will do in terms of creating an illusion you actually know what you're doing when in doubt add decimals the intimidation factor is amazing Excel you can go 10 decimals 15 20 25 just keep doing it at some point people step back and say I'm not asking this guy questions he's got the 25th decimal point nailed down and if you're on the other side of the table remember this when you see more decimals what's a guy doing he's basically trying to intimidate you hey so round It Up Round it to the nearest billion might as well right so you know so bias comes from lots of different places but there is bias now I'm not even going to go through this page I have this weakness of trying to fit as much as I can into a single page my objective in life actually is to take my entire class and see if I can fit into one slide one of these days I'm going to be able to pull it off but if you're doing valuation keep this page this is my template for how you can introduce bias into your valuations and there are actually ways you can introduce one of the ways you can separate people who know how to do valuation people who don't is if you know how to do valuation you can hide your bias better if you don't know how to do valuation your bias is obvious you do stupid things like assume a 6% growth in perpetuity okay there are ways of hiding bias so if you get a chance go through this this is my template for building bias into valuation but it's also a template you can use to detect bias when you have it in valuations so let's talk some numbers some of I was asked to talk about Facebook so might as well talk about Facebook Facebook's been this one of the two stories of the year right one is Apple the other was Facebook so this was on May 17th of 2012 the day before the IPO now I've always believed that you're going to do valuation you have to throw the numbers out there I've never believed on the one hand the other hand hedging so basically I said I'm going to do evaluation I'm going to throw it out there could I be wrong of course what percent of the time 100% of the time but one of the nice things about doing valuation is it's okay to be wrong because all you have to do is be less wrong than the next guy and they all screw up so muchy I'm okay with this so the day before the IPO and as you know that 4 P.M that day the IPO price the offering price was set at 38 I did a valuation of Facebook so this is my valuation the day before and I post in my blog saying I know that Morgan Stanley has access to far more information than I do and I'm sure they have a really good basis for pricing Facebook and I was very clear that they were pricing Facebook not valuing Facebook this is my estimate value and a value that I came up with $25 did I make assumptions God did I make assumptions every number there is an assumption Revenue growth margins Etc now to show you though how much that value is a function of assumptions I'm going to show you two more valuations just if you if I'd been working at Morgan Stanley thank God i' wasn't but but if I had and my job had been to justify the $38 price I could have tweaked this valuation and you wouldn't even have Notice Me tweaking it I and and I've kind of listed my big assumptions it basically boils down to the fact that I made a small change in my assumptions about margins over time and I made a small change in my assumptions about return on Capital and perpetuity and if you don't get the NR valuation you probably don't even notice the changes happening in the background and with those two changes made the value per share that I get is $39. 32 there you can justify the $38 price now if you'd come to me with a different agenda and said I'd like to sell short on Facebook I want a $15 value no problem again with a little tweaking I can come up with $14 what I'm trying to say is if you have a valuation model and you have real buyas there are ways in which you can make make that model sing the tune you want it to sing it's bi aware so if you're on the other side of the table you can be intimidated by this big model but you want to focus in on the big drivers of value and if you focus on the big drivers you can sometimes if you're lucky spot that bias when it shows up in your valuations so let's let's let's talk about the other way in which people do valuation which is pricing okay there are three steps to pricing you start and you pick a multiple right which one whichever one works for you we act as if multiples come from Heaven they're not you as an analyst decide whether to use P ratios EV to AB EV to sales e to invested Capital whatever so here's my job if I'm an analyst valuing a company and I have to come up with a number I try seven multiples I come up with seven different values which one am I going to use whichever helps me tell my story second step in the process I have to pick comparables I have to decide what I mean by comparable am I going to look globally just locally am I going to Define it narrowly by sector or broadly by by business so essentially I get to decide what multiple to use I decide to pick the comparables to use if you let me pick the multiple and you let me choose the comparables you've seeded the entire Battleground to me because I'm going to find a way to justify whatever number I want you think real estate in Mumbai is expensive all I need to do is compare to real estate in in Tokyo just find something even more overpriced and say you know what that's cheap you let me pick the multiple and you let me pick the comparables you've essentially let me decide what the final number is so be a disturber when somebody uses e to say well that's nice what try e to book in other words you want to get them off the game because if you let them set the set the rules of the game you've lost control of the process if they show you 15 companies those are my comparables ask did you start with 15 or did you start with 25 they're going to lie and say I started with 15 so don't lie to me show me the 10 companies you threw out not the 15 companies you left in because that's going to tell me a lot more about your valuation than just what you see in the output so it's it's Again part of the process and something to think about whenever you think about looking at somebody else's valuation so let's talk about dealing with Buy here the there there are unhealthy ways of doing this and healthy ways of doing here are the unhealthy ways the first is the analyst says look I'm not biased I use only numbers as if that's a defense against being biased in fact this has become the basis for anybody doing valuation I'm not bi look it's all numbers they came out of the accounting statements you can be all numbers and still be biased you saw my Facebook valuation there were all numbers all three valuations but values ranging from $14 all the way to $39 the second is for some reason I've gotten roped into doing these talks at CFA these acronyms gone crazy all these valuation certificates every one of them has this professional statement that says our professional thing is we are we are going to be unbiased as of putting that down on paper will make you unbiased if it only that simple and the third and this is I think a game we've seen played is I've got this fous opinion signed off who the heck cares you know there's a fairness opinion for the worst deals done in history the time owner AOL deal there is a fairness opinion somewhere in somebody's Pho saying that's fair it doesn't matter see it's not even it it doesn't even matter because I think it basically reflects the fact that fairness here means nothing in fact um I was asked to talk to a group of CPAs why was I asked to do that maybe to T me or something the CPA and they were the rule writers and they wanted me to talk about fair value accounting and I put up my first slide and then knew where I was coming from I said fair value accounting oxymoron make up your mind do you want to do valuation do you want to do accounting don't do both because you do neither right because the notion of fair value in the hands of accountants is a little meaningless because you can write all the rules you want but there's this bi that that feeds into the process it's a make work for accountants now that's basically what it is because you have fair value accounting they need to revisit you every year to fair value everything you have that's a lot of money to be made right so that's the first issue is with Buyers a lot of people deny they just act like it's not there so here's what I'd propose I don't think there's there's much chance of it happening in the short term is we need to start thinking about building processes where there's less bias there's no way you can build a proc process with no bias it's impossible to do less bias over more bias let me give you a couple examples equity research is going to be biased the way it's structured is you have an analyst who tracks 15 companies for the rest of his life you can ask him to be honest about these companies because if he's honest about these companies they're going to cut him off that's the rest of his life the nature of the process is an equity research analyst cannot afford to be unbiased because of the way we've structured this process m&a of course it's going to be biased cuz what I'm asking I'm going to the dealmaker and ask him does this deal make sense the analogy I use is like asking a plastic surgeon is there something wrong with my face what do you what do you expect to hear you're perfect I mean if he says you're perfect he has nothing to do right you go to a dealmaker and say does this deal make sense he has one of two answers he can give you he can give you the honest answer saying don't do this deal it's a terrible deal in which case what does he get the undying gratitude of my stockholders but trying paying bonuses with that or you can say the deal makes sense in which case it's 50 million 100 million 1 15 the the process is skewed towards saying yes on the deal so we need to start thinking about processes because until we fix the processes the valuations are not going to get fixed second let's be honest at least with ourselves don't put it down on paper you could get sued subpoena whatever but be honest maybe implicitly when you're late at night with just your maybe not even with your spouse around cuz who knows what you know she might be witness for the prosecution or something in this Cas so maybe when you're in the bathroom look in the mirror and say today I made up some numbers I don't think Facebook is really worth $38 I will tell nobody but myself right but let's stop lying to ourselves about this hey I'm honest I'm objective I'm just saying that not telling the truth as I see it in fact in statistics the branch of Statistics called beijan statistics where you're supposed to state to priors before you tell me what you know so you run an experiment tell me what you thought you were going to find before you run the experiment maybe we should have a similar practice in equity research the analyst says up front I really love this company now let me show go through the numbers and see if in fact I can show you my love right it's it's it'll be good to know where people come from but as I said it might be easy to do that if you know which direction the bias comes from and be transparent about your motives so I can't force you to be honest but I can at least learn about what your what access you have to grind because that will allow me as a person using your valuation to decide how much I will trust your valuation let's talk about uncertainty if you think about valuation you think about all these inputs cash flows growth rates and every one of the inputs you face uncertainty so if you if the generically I think about four basic questions that need to answer to value of company what are the cash flows you get from existing assets what is the value that you're going to create from growth in the future don't notice how I frame that question I didn't ask what's your growth rate going to be in the future but what value you create from that growth right how risky are you as a business and when will you be a mature company all of the inputs and valuation basically boil down to those four inputs and every single one of those inputs you estimate with uncertainty but not every company is equally difficult to Value right in fact I'm going to show you three valuations the first was the valuation of 3M companies been around a long long time does boring stuff but boring very profitable stuff like that Post-it pad that a stroke of Genius right the guy who invented that should get a Nobel Prize because think of how many people use it but it's incredibly priv it's a company it's been around a long time you know exactly what it does margins are stable it does pretty much the same thing and this was a valuation in the good old days when develop markets would develop markets and Emerging Markets are Emerging Markets every market now is an emerging market so those good old days are pretty much gone right so this is valuation in 2017 we were valuing a stable company in a in a fairly stable economy so I did the valuation the number it's true I made estimates but I felt with each estimate if you think about a margin for error I felt pretty comfortable the Distribution on the inputs was much tighter the value that I got was $83 the stock was trading at 70 but I mean that's not what I'm going to focus on here's a second valuation valuation of TARTA Motors in 2010 a company that that that has history you can see what's happening but couple of things are kind of rocking the boat here one is that in the year prior to this valuation you had two big changes one was the internal investment in the Nano kind of changing the mix and the other was the acquisition of Jaguar Land Rover which is changing the mix as well in an economy which has so more macro uncertainty associated with more uncertainty here because again a combination of what the company had done in the new past and what the economy was doing and the third valuation I'm going to show you is a valuation of Amazon in early 2000 that I fogged to death over time because in a sense it's valuation in January 2000 for those of you who don't remember was when the NASDAQ H 5000 the peak of the do com boom tremendous uncertainty about the future about the company not so much about the economy what I was what I'm trying to highlight with all three valuations is when you think about uncertainty in valuation comes from lots of different places it can be estimation uncertainty or economic uncertainty people mix up the two estimation uncertainty is mistakes you're making an estimation economic uncertainty comes from the outside and the reason I emphasize that is there's this illusion that if you spend more time doing valuation you can make uncertainty go away I'll give you an experiment try valuing a Greek company you can spend from now through eternity refining your valuation and guess what you're still going to face uncertainty some uncertainty economic uncertainty you can't make go away by building bigger models so it's estimation uncertainty versus economic there's micro uncertainty versus macro uncertainty what am I talking about with Amazon there's a lot of micro uncertainty right I don't know what the margins be What markets are going to sell what the revenues will be the cash flows will be with TARTA Motors there's a lot of macro uncertainty you're saying so what again the way you deal with it is very different micro uncertainties you try to refine through your cash flows macro uncertainties you try to capture in your discount right and there is discrete uncertainty versus discontinuous uncertainty what what am I talking about let's say you value Venezuelan company or a Russian company you do all the standards stuff you project the cash flows you discount them you come up with the value and then I say oh by the way you might have forgotten something very important if your company looks really really really good what might happen Hugo Chavez on his deathbed might say hey I've nationalized the company that's uncertainty it's it's a very different kind of uncertainty the uncertainties were very good at dealing with evaluation tend to be you know continuous uncertainties basically that happen all the time that's what he built through the discount rates discrete uncertainties where you could basically you could either go bankrupt or you can't you get nationalized you don't are much more difficult to deal with so one thing I would suggest when you do valuation and you find yourself faced with uncertainty step back and think about what type of uncertainty is bothering you because the way you respond is going to be different depending on the uncertainty right so let me make some suggestions on uncertainty he so if you look at the four valuations we've seen you can see that all of them um reflect uncertainty but i' probably feel more certain about the 3M valuation in 2007 than I would about the Facebook or the Amazon valuations or the T Motors Val so if you think about uncertainty not all companies are equally difficult or easy to Value so if I ask you to pick a company to Value which would you rather value 3M or Facebook but be careful why do we do valuation because we want to make money of those valuations right so you pick a company that's easy to Value congratulations you're going to be done xon mobile anybody can value xon mobile including my 14y old it's not a difficult company to Value the pay off to doing valuation is not how uncertain you are about the value of a company but how uncertain you are about a value of a company relative to other people valuing the company and here's one of the big ironies in valuation the payoff to doing valuation is greatest with companies you feel more most uncertain about the future the payoff to doing valuation is greater when there's a lot of macro uncertainty than when there isn't because after every crisis I get asked this question does it make sense to actually do valuation in a market like this one and I say it's exactly in a market like this one that you should do valuation because most people give up so here are unhealthy ways of dealing with uncertainty they're human nature right first thing is what uncertainty I don't see any uncertainty and the analogy I would give is you're driving down a highway your GPS is failed you come to a fork in the road you have not sure which way to go I know what I'll do I'll just stop paralyze I don't know what to do now or you can say I always go right I don't know why but always go right that's mental accounting you're basically using a rule that has no basis in reality you just make up these rules or you turn to your wife and say which way should we go that way you can blame her that's Outsourcing the uncertainty right I mean in other words we don't deal with uncertainty in any healthy way it's not like any of these actions are going to make it a better choice but we do it all the time and people do that in valuation so I won't get a chance to go through this but about um my the presentation for this is on my website I actually at the aimr conference about two months ago talked about 10 things you can do to deal with uncertainty and I'll very quickly list them so you can at least think about healthy ways to bring in uncertainty first remember when faced with uncertainly less is more when I value difficult to Value companies the number of line items I estimate gets smaller rather than larger it goes against human tendency which is to add more detail when you feel UNC have less detail aggregate don't disaggregate build an internal checks within your own valuation in other words make sure your valuation is not at war with itself okay third when you make assumptions about something make sure there's something offsetting it so for instance if you think inflation is going to be low that's going to help you on your interest rates and your discount rates but it's also going to hurt you when you talk about growth so if you have offsetting effects it has a much lesser impact on valuation drawn first principles you can't make up rules as you go along so if you tell me your company is going to have an 80% return on Capital forever I have a problem with that because that's really not an estimation issue it's a question of has any company ever been able to pull that off what kind of competitive Advantage is this that you have that allows you to do it I use the market as a crutch all the time CU even if I think markets make mistakes the one thing I cannot afford to do is ignore what markets are telling me so even if you have very little respect for the market price you think investors are crazy there's there's a there's a method to the to that Madness there's something you can always extract so when I know one of the key inputs in valuation is an equity risk premium and if I ask you where you getting your Equity risk premium the answer I get is you've outsourced it right you looked it up on abson you looked it up on my website you blame me the motor and got it wrong it's not my problem but there is an implied Equity risk premium in the market you can back it out of the market and you can use it so use the data in the market when you're faced with uncertainty as a way of kind of getting over use LW of large numbers Baars me out every single time what am I talking about when I have to estimate the margin for a company I look at their margins last year that's one one point or I can look at their average margins over time the average margins across the sector averaging might seem like a simplistic thing to do but it's amazing how much power there is to the law of large numbers yesterday I finished my update for the 2012 data sets you know every start of every year I update my data and one of the reasons I compute these averages not because I'm a Mother Theresa wants to give away the data but those are the numbers I'll be using for the rest of the year when I'm doing valuation so if you ask me to value Steel company I can look at its existing margin but I'm always going to measure it up against what the rest of the sector is going to look like what those averages look like don't let the discount rate become the receptacle for everything that's scaring you there's a Temptation there to go push up the discount rate I'm scared I'm going to push it up okay it's not a device meant for everything that you're afraid of so don't let it be the placees I'm there's liquidity so let me push that discount rate up when you're faced with uncertainty don't avoid it right what I'm talking about is there are tools and techniques out there for bringing uncertainty into valuations I use crystal ball do any of you use crystal ball in your value it's it's great looking output even if it's garbage in garbage out you can turn out some really goodlooking garbage okay but basically if you use it right though what does crystal ball allow you to do it allows you to introduce uncertainty about your inputs right so when I ask you what the revenue growth rate is rather than give me a number you give me a distribution which is reality 25 years ago I'd have to pay somebody tens of thousands of dollars to run simulations for me now with a few hundred you can install it on your laptop run it off Excel and you're there and finally don't look for precision as I said you're going to be wrong 100% of the time so what be ready to be I mean the key is you're going to make mistakes if you say I want a model that doesn't make mistakes you never going to do valuation so if you get a chance kind of go through those pages you're getting these pages I go through right basically you know actually I have to tell you that I just um the the page that's in your slides is June 2012 the June January 2013 country risk premiums are up this is the most upmost downloaded data set on my website it gets used in the strangest places I get emails from places I remember getting a email from the New Zealand milk board saying we're using your country risk premiums to set prices for New Zealand Farmers why no I don't know who are they selling to China I don't know maybe that's what they were doing you know and usually they how do you come up with these numbers and I have a little PDF file and what because there's no rocket science here but I do get some emails that are a little way over the top I'll give you my favorite or least favorite of all time it's about two years ago in 2009 I get an email from a business person in Lebanon I'm very excited to get an email about valuation in the Middle East I don't get that many questions so open up the email and this is how it begins you have destroyed Lebanon so what how did I do that I can see myself destroying lonstein or Luxembourg or one of those quasa European countries but Lebanon after 25 years of Civil War I'm the guy who destroyed this so I keep reading what did I do so it turns out that he was a Lebanese business person whose business had been a praise for value and as appraiser had pulled the risk premium from my website to Value the company and if you look at Lebanon it's right there right now it's 12% but that was 14 1.5% in 2009 you see why this guy was really pissed off you build in a 14 and2 % risk premium you have high discount rate you have a high discount rate you have a low value so this guy thought he got jipped on his valuation was all my fault my first response was not my problem take it up with your appraiser and then I remember the email was from Lebanon and I decided that discretion was the better part of valor here so I look for somebody else to blame you know how I get these premiums what do I start with I start with the Sovereign rating for the country why there's no way I can do research on these countries to figure out how risk so first mistake I make is I trust the ratings agencies right but it's cheap it's free actually so I said it's not my fault it's Moody's fault and by the way if you want their address here it is put into your GPS park your car over there don't do anything near my building I have nothing to do with this so if you get a chance kind of go through these risk premiums but there's no intellectual Firepower B any of these numbers don't do anything rash with them I'm pretty much running out of time so let me go to the last part of valuation oh there's a nice looking output I talked about the only thing about about about these simulations is before you do it I know you those those of you took I mean you took your statistic class and I know you sold that book back the minute the class ended said Thank God this class is over I don't have to ever look at this book again go and buy it back because the skill set that I see most lacking in valueable ation is basic statistics okay so if you're going to use crystal ball at least get a sense of what the difference is between the distributions because if you don't guess which distribution you end up using for everything what's the only distribution we have any connection to after we leave that statistics class normal I'm normal so I'll just pick the normal distribution right nothing is exponential nothing is uniform everything is normal so at least do that homework on different distributions because there is this incredible payoff you will get which is you can then take your uncertainty face up to it put it in there and let the output reflect what you find so the Facebook valuation for instance in addition to doing my valuation I said you know what I could be wrong this is how wrong I could be here's the distribution is there a chance that $38 is a fair price absolutely there's a 31% chance it's it you could buy the stock and it still be cheap I could I should say 31.5 35% the the third decimal point again creates a delusion that I'm actually talking about something that's more precise than actually is but deal with it which brings me to the last piece in valuation when you look at complexity in valuation some of it is coming from the outside companies are getting more complex they're multinationals multi I call these companies octopuses basically they have you know tentacles all over the place making them more difficult to Value data is getting more accessible that's a good thing right but it's also making valuations more complex Capital IQ you can get 350 line items about it every company on the face of the Earth you wouldn't have know 25 years ago you were digging through an annual report looking for those numbers that kind of constrain how much data you could use and you can build bigger and better models so much more easily than you used to be able to so that's all feeding into making models more complex you also have complexity fed from below analysts sometimes build complexity because it intimidates and if you're objective in doing evaluation to make sure nobody has any idea what's going on you build a more complex model and and you're also getting complexity coming from legal accounting so all of this is feeding into making models bigger okay I'm going to stop with this slide but I'm going to talk about the complexity and what it does first is it does two things one is it creates input fatigue know what input fatigue is I'll describe it and those of you do valuation will probably relate very quickly to it it'll hit you around 11:00 on a Saturday night you've been working all week in fact you've been you're working through 11:00 on Saturday you're exhausted you get ready to turn off your computer to go home your managing director comes and then throws a 10K on your desk so I want this company valued on my desk first thing tomorrow morning why Sunday morning it's his trial by fire he wants to see how much you really want this job now part of you wants to exercise your option to abandon I won't describe it but it's a deeply satisfying option to exercise but then there are deeply dissatisfying moments that follow usually takes a form of the 10K and throwing it back in his face saying do it yourself that's a deeply satisfying moment but then everything after that is downhill so you think about it then you hold back you remember your car payment your house and all that stuff and then you decide to do the valuation so you sit there starting entering the numbers you have this model that's been built by this in-house team and God help you there's a team in-house that builds models for you it's geek city right you hire people they go always go to the basement for some reason they don't see the light they're like vampires hey they love building models and macros on top of macros and bigger and bigger the model inputs come and they never leave so model starts with 10 then 15 30 50 8 85 inputs to Value one company so you sit down you start entering the numbers you get to the 10th input the clock strikes midnight you're not Cinderella you wish you were then you look down there's 75 more inputs to go your stomach drops then you look at the 11th input says what was the inventory 5 years ago part of you wants to get up and look it up but that part's too exhausted to get out of the chair the other part prompts enter a random number let's move on and it's amazing how quickly they all roll out right and the scary thing is when the output comes out it all looks the same that input you toiled hours over random numbers all melt together that's input fatigue the other is the model becomes a black box I still remember a conversation I had with an equity research analyst from JP Morgan I think about 20 years ago about ay company had a buy recommendation on a company I was familiar with it had a Target price of 85 said how do you come up with such a high value what do you like about the company he said I didn't do it I said what do you mean you didn't do it your come your names on the report it says $85 he said I didn't do it I said who did it he said value MAAC did it I said who the heck is value Mac says that's our in-house valuation model I said what did you do sneak into your office in the middle of the night value the company and leave it on your desk but you know what he was trying to say look I fed the numbers into this model something happened there $85 popped out in fact look for these words in any valuation report if it says the model valued the company at you know what the analyst is saying right he says saying don't blame me I just work for the model it asked me for inputs I feed it in it asked me to go get a cup of coffee I come back it's left a valuation on my desk okay models don't value companies you and I do and I think when you think about complexity the healthiest thing to do is kind of strip complexity away it it goes counter because we to build bigger models use more data less is more I've never done a valuation with more than eight line items ever I've never ever broken down working capital into you know how you can break working inventory if you can predict accounts receivable in the year 2030 all the more power to you I have no chance of doing it why even try so keep it simple because as I said when you think about valuation those are the three things that get in the way of good valuations bias uncertainty and complexity and if you can learn to handle them the rest is just mechanics so that's about it so let me um let me introduce the panel that's that we're going to talk about valuation

Channel: NYU Stern

Views: 143,120

Keywords: new york university, nyu stern, stern school

Id: -lL5qj_h1RE

Length: 47min 51sec (2871 seconds)

Published: Wed Jun 05 2013