Session 7: Measuring Relative Risk - Betas and Alternatives

Leon Constantine and France you you're in both my classes anybody else in both in both my classes he for four of you the last three or

four sessions look like they're paralleling you know enjoy the moment while it lasts because you know the risk free rate in the equity risk premium and the beta discussion for the most part of

the same and everything so today we will kind of diverge so easy days of listening to the same thing twice are done sir just just let you know

statement okay okay welcome back from your long weekend I hope you had a time to pick a company analyze it I hope you enjoyed yourself

and did that as well just a reminder your first quiz is a week and a half from today it's two weeks from Monday so a week and a half from today on March

3rd I know again for those of you in both my Corporate Finance in my evaluation class you'll have backto back quizzes so you know it is what it is so

yeah um it'll cover everything we do through next Wednesday obviously so it'll be primarily discount rates and cash flows I'll tell you when in the slides we'll stop until that's what

we're covered but everything you need for the quiz is already up online the review session pass quizzes and my advice is if you want to prepare for the

quiz do more problems the more you get your hands dirty working through problems not just check the problem the solution working through proms the better off you're going to be so today's

class we're going to do a shift away from discount we're going to talk about cost of debt and cost of capital we're going to complete that discussion but then we're going to start on what I think is the most critical part of

valuation assessing cash flows okay so I'm going to start with a few questions we will arrive you know we will talk about more today but I want to get at least set the

table remind me again when your valuation is due the last day of class you know what day what day is that May 5th right and you

have to put a buy or a sell recommendation as of that day right I will give you some leeway and let you use a Friday price from you know May 2nd or something so you're comparing the

price from May 2nd to a valuation that technically should be of May 2nd but I'll cut you again some slack it might be from April 30th or April 15th but you want to get

your values close to price because you're Deion has to be in real time now if you think about the inputs that go to valuation some of those inputs get updated every second of every

day like what what are some of the inputs in your valuation which will be every day every minute of every day the t-bond rate you mean the equity risk premium I'll have a May 1st Equity risk premum if you feel the urge to update it

hopefully it won't be that different from April April 1st your market price will get updated every moment of every day but a big chunk of your information

comes from from financial statements right and they don't get updated every moment of every day or even every day or even every week or even every month in

the US they get updated once every three months and not even at the end of the month it'll often be six weeks after the month ends in a long and Tor torturous

way I'm setting up the first question do you want the most updated information you can get for things like T Bond rates and market price will be today's data but for things like accounting you are

stuck with whatever your most recent financials look like so let's say you're valuing a company today not even May 5th you're valuing it

today other what what's the company of value on the Swiss shoe company right what do you what's your last annual report look like which which your that

what year but where what yours annual report you have access to so when you go on the website you have the 2024 numbers on already updated or not because December

was the ending right no no that's not the way it works right you can't say because of your end is December the annual report now is going to be the 2024 because it takes time for the annual reports to come out I wouldn't be

surprised if you went to the website and you check for the last annual report the most recent annual report is the 2023 annual report okay for the moment at least it will change as you go through

mine is last March and then I detect you okay so we'll talk about that so your annual report is so already you can see the problem right the last annual report is of might be of 2023 it might be March

of 2024 it might be June of 2024 Apple has this very strange fiscal year end in September of each year and you doing evaluation in March or February or

April so but you can say that's my last annual report I can I will stick with that so that's the first choice the second is Gabriel mentioned that he had the last quarters numbers right can I

just multiply that by four it's more updated right but what's the problem multiplying last quarters numbers by four you take a retail firm and you take the most recent quarter God only knows

what you get right there seasonality so that's off the table so you have the last annual report last quarter multiplied by four that's not going to work what can I add the last

four quarters together fact what is that called when we add the last four quarters together it's called trailing 12 months last 12 months and in fact it's often LTM TTM

you will see this often in financials last four quarters or you can take if you have numbers through September of 2024 let's say the first nine months you can take

those and maybe fill in the last quarters because there already analyst estimates for what the 2024 numbers the last quarter will look like and create your own version of

2021 let's rule out what you should not use the last quar multip M by four you should not use because of seasonality what about the last annual report even

if it's a 2023 report Julie what kinds of companies would you be able to get away with using the 2023 report companies were not much as say

you take Coca-Cola 2023 2024 you can be 2021 I could probably get away with it a ramco what's different about 2024 oil price might have changed if your company

is a mature company you can get away using old financials and not pay a price for it but you take Nvidia you take the 2023 report and you compared to the last

four quarters it's like I mean you're looking at a very different company my rule is never use the last annual report go with trailing 12mon

data and in fact staying on the trailing 12- Monon data we talk about you know when you say trailing 12 month dat it sounds like a lot of work right because what's your mind thinking about four quarters of data I've got to get four

quartly reports but you actually don't have to if you have 110k the case of you have a March 23rd 2020 and

110q I'm going to show you how with 110k and 110 Q you can create a trailing 12 months it's not rocket sense you can add basically you can net out the numbers and come up with the trailing 12 month

numbers so the key point is get as updated as you can on the information some will get updated every minute of every day some will get updated every quarter and some might you might be stuck with the annual report data I'll

tell you in the US while I use trailing 12mth data for most accounting numbers there are things like stock options outstanding that will show up only in the annual report nothing I can do about it I'm

going to take that as most updated number and work with it the second question this is more of a generic question once we get earnings for a company we need to subtract out what

it's reinvesting and one big measure for reinvesting is what a company is putting into Capital expenditures I'll give you the definition of capital expenditures that my accounting professor gave me and then

I'm going to list out some items and you tell me whether you would think of them as capital expenditures or not I was told that if you have an expend expenditure where you have benefits over

many years it's a capital expenditure right keep that definition your mind because I'm going to read off a list of items and I want you to use that definition not what you see in an accounting statement or financial

statement to tell whether it's capex it's expense that creates benefits over many years building a manufacturing plant unless you have a really short life cycle product right if your product

just last one year you might actually argue that but in most cases plant you going what about R&D expenses super care PEX right what

company in its right mind does R&D expecting to get a benefit this year if you're lucky maybe two years maybe 5 years if you're a technology firm and if you're a pharmaceu itical firm you might

have to wait 8 10 15 years Nova nois is living off R&D it did in 2013 and 14 or maybe even earlier for OIC and

wovi what about advertising by a consumer product company there are two reasons companies advertise right one is to get you to buy products today that's an operating

expense but if you watch the Super Bowl ads they were not about getting you to buy stuff today they were about building brand name and you know so if you saw an

ad for on it's really about not getting you to go buy on it make you aware of this company the brand name and if that is your biggest asset let's face it for a company like Coca-Cola and on and Nike

your biggest asset may be your brand name the money you spend building that brand name is really capital expenditure by now you see the pattern

here right much of what we think of as capex in a financial statement is oldtime manufacturing capex so when you go to the statement of cash flows for your company look under capex you're not

getting R&D you're not getting recruiting and training expenses for a consulting firm which is basically their biggest asset is building up human capital you're getting old-fashioned land building equipment

machinery which effectively also means that when you're valuing a technology company in fact most companies you can't trust the capex you have to come up with the with

a measure of capex that's far broader than the accounting capex because you got to C if you don't do that everything is going to get misstated

question yeah that's part of it too right so any money you so think long that's basically go back to the core definition and say an accountants are not consistent with their own core

definitions right so that's what we're doing is we're holding them accountable saying if that's your definition of capex I'm going I mean if you're looking

at a company like Uber or Airbnb what's the biggest capex you think at those companies software is part of it Cal

customer acquisition costs right how do they get them by giving you freebies to try the platform that will show up as a cost but to them this is how you build up your

business is having more users more writers more subscribers so already you can see that the companies of today the capex you see on the statement of cash

flows doesn't capture what they're actually putting in for future growth so we're going to come back and address all of those issues so let me or one final

question I want to kind of draw a distinction between two different tax rates we're going to go back and forth in this class the first is called an effective tax

rate what what's an effective tax rate think of it is you take total taxes paid by a company and divide by total taxable income you have an effective tax rate so

you pay 200 million in taxes on a billion dollars in income 200 divide by a billion is 20% that's your effective tax rate the marginal tax rate so the effective tax rate you will find in the

financial statements and even if you don't find it go to the income statement take taxes paid divide by taxable income you got the effective tax rate the marginal tax rate is the tax rate you

pay on the next dollar of income or the last dollar of income you won't find it in company financials this you find in the tax code let's see how familiar you

are with the Us corporate tax code what's a federal corporate tax rate in the US right now 21 right what was it in 2017 before

the tax it was 35% so 35% to 21% and here's the weird thing us tax laws because of the way you got

to get through the Senate and you have to get passed through with a filibuster you got to reconcile because of that tax codes come with finite lives so the 2017

tax code is expected to completely go away at the end of 2025 it falls fact it's there's a cliff you'll approach and if you don't fix it it reverts back it's

a weirdest stuff you've ever seen you know who does this but this is the way it works so it could very well be 35% so who won the election this year might

have a big effect on evaluation because you know to the extent that you know the tax code gets extended when you value companies today you might use a 21% starting point but if it had been a

different outcome and you said that tax code is not going to get renewed you would have to actually adjust your marginal tax rate going forward because you expect it to be you know so this

year actually the question is will the marginal tax rate get lower rather than higher because there is stock at least of the reconciliation this year lowering the corporate tax rate from 21% to a

lower number for at least a subset of companies Callum question about all the capex and taxes because part of the problem like I I agree with you that like there are things that are

investments in your company but some of them are cash tax deductible some of them are right so when we're looking at like getting our free tax FL we count expense like advertising I think you're

raising a very interesting question the tax law is the tax law you're not going to change it know it's got nothing to do with good sense bad sense it's what's tax deductible there's no reason why we

can't treat customer acquisition cost is capex and not a bring it's easy to capture the tax benefit from it right so I will actually have my cake needed to I

will move the items into capex if I feel they're being mistreated but I'll still compute the tax is based on the existing tax code that's where that way I don't leave any of the tax benefits behind

separate exactly and it's actually very easy to do I'll give you a shortcut to do it because you have a good point which is we're moving it all these items from operating to Capital expenses but the tax law still allows me to subtract

R&D as an operating expense I'm going to take it I'm not going to go to the IRS and say look can I spread it out over 5 years it doesn't make sense for me to do it so I think you're absolutely right we

have to know tie up some loose Sense on taxes so effective tax rates marinal tax rates right which one do you think is going to be higher for most

companies the marginal and why well I want to start digging on this because this is going to tell you what to do Jacob why tax would have any like tax savings from last year losses or

something one is carry forward so that's one thing and the other is you have very clever tax lawyers you might be able to defer taxes it might also be with us

company we have foreign income after 2017 and this is more puts us more in line with the rest of the world if you have income in a country with a 15%

corporate tax rate you pay the 15% corporate tax rate and you're done there's no extra tax due that didn't used to be the case before 2017 already you can see a possibility of of

effective tax Apple's effective tax rate is 12% some of it is no no losses carried forward they haven't lost money in Forever is tax tax defer a lot of it is

income in locations where they have a much lower corporate tax rate okay that's not going to go away you know as so one of the things we're going to talk about is what do you do when the

effective tax rate is less than the marginal tax rate valuation do you leave it at that lower number do you move it towards the marginal tax rate and I'm laying the foundations for answering that question and you're going to see

the answers are going to be different for different companies for Apple we might leave it at 15% below the 21 % because if they continue to be a

global company getting 50% of the revenues outside it's never going to catch up but if it's deferred taxes we have a problem right because eventually those tax are going to come

to you there I'm going to be more likely to move from 15 to 21% so let those two things kind of move around and think about the difference because you're going to see it play out in valuation in

different ways when we do cost to debt as opposed to cash flows but those are last year the average effective tax rate for US companies was

18% you know the average effective tax rate in 2016 before the tax code was it was

19.2% you say but the marginal tax rate went from 35 to 21% you know we talk a lot about the marginal tax rate but companies when you

have marginal tax rate of 35% were playing all kinds of the games so the actual net loss in tax revenue from going from 35 to 21% was minuscule if

you look at the total corporate taxes paid in 2018 as opposed to 2016 right so the effective tax rate captures all of those actions that companies take to

keep their tax rates low the marginal tax rate whatever in the call incidentally for the US the federal corporate tax rate is 21% the reason I emphasize that is if

you step if you as a company you have revenues in California in addition to the federal corporate tax rate you have to pay state taxes and if you're in New York City local taxes so by the time you

add all of that up the marginal tax rate in the US is closer to 25% which happens to be roughly where the marginal tax rate is for much of the rest of the

world so the 2017 tax code actually brought the tax marginal tax in the US closer to where it is in Europe for most of Europe I think Germany's at 29 no

Ireland is at 12 so those are the limits in Europe but much of Europe is around 25% but it's going to come into play as we talk about cost of debt and why companies borrow money and what the

benefits of thatt are right so let me turn the lights on for whatever reason people let me go back to where I left you on where we left off in the

packet so back in the valuation packet for those of you tracking which packet so that was the start of the class let's go back to where we were we were talking about building up to a cost of equity

and we had the first two ingredients nailed down right at least partially Jessica remind me again if I asked you what a risk-free is how would you answer the

question before you do that ask me in what currency right you are quick to jump to US dollars right so first thing with risk free rates is don't start the conversation until you have currency in

Elda right it's in US Dollars I'm going to use a treasury rate shortterm or long term not in valuation right because everything is longterm you're looking at cash flow so you going to go to the long-term treasury rate and you done

other currencies it's a little bit of a mess because you might have no default free entity but the risk-free rate is always currency driven it's got nothing to do with where your company's Incorporated it's nothing to do with

your frame of reference or what what currency the company borrows at you make a choice Jacob help me on the equity risk premium to come up with the equity risk premium for a

company start with the equity risk premium in the country but first what country whatever country so before I do that what's the first thing thing I have

to ask where do you do business right so don't jump to the country because then you're going to give it the equity risk the country in which it's Incorporated but as we said you can be an Indian company with 90% of your revenues in the

US so with the equity risk you start with where the business operates and if you want to use revenues that's fine you want to use production for natural resource company you might do that and you then you build up to a weighted

average of the equity risk free risk free rate Equity risk free what's the last ingredient to get to my costomer Equity I need a beta okay let's talk

about betas because there's an immense amount of noise around betas especially after you've taken a finance class because in a finance class what are you

taught betas come from regressions you're under regression of returns in the stock against returns in the market index the slope of the line is the beta and statistically you can see the

rationale beta measures risk related to the market and beta captures therefore that correlation or co-variance with the market it's a terrible way to think

about betas in my view first because it makes it a statistical number right when people ask where the bet is come from they come from a regression and if you remember your

statistics when you run a regression you get a slope If This Were A statistics class you're trying to put in standard errors right below the slope how uncertain you about the slope a practice

we very conveniently forget after that statis just like in the equity risk premium the promet regression bait is at threefold first they're noisy you've heard that before Equity risk premiums

you hear that now with betas they're noisy because they come with a big standard eror second they're backward looking by definition because to get a beta I've got to run a regression of not the next five years of returns because I

don't have them but the last five years of returns I did that for Disney remember think of how much Disney has changed over the last 5 years there was no Disney streaming at the start and by the end you got this big chunk of money

being spent on Disney streaming companies change and the beta you use will reflect the business or businesses your company was in during the period of the regression and we're going to talk a

little bit about how debt plays out here but debt can make your beta higher because it increases Equity risk and do companies debt policies change over time absolutely they can go from no debt lots

of debt or lots of debt to no debt and that can change your bait as well so I'm I'm going to dig a grve for regression betas and bury them the way I'm going to do this is by taking you

through a series of beta examples to explain why it's so dangerous to trust a beta 95% of the world if you ask them where the beta comes from they go to Bloomberg they

print off a beta page if you're an investment bank there's a beta for my company I'm going to give you the ammunition to take that beta page apart so let me

start with a company that most of you must might have heard about GoPro remember GoPro in its Heyday what did it do it provided cameras for oversharing

overactive people I think basically you went on a 3our hike you put a GoPro on your head you recorded the whole thing you carried it live as if people cared

watching you for three hours and a hike but it had its glorious moment right big Growth Company so I know ran the regression for I went to Bloomberg and the nice thing about Bloomberg and I

gave you the instructions on how to do this once you pick a company you type in beta you get a page like this there's a regression beta page for Bloom and it looks great there is my Beta

1.64 the adjusted beta you can ignore it's just a raw beta push towards one but basically this is a regression beta you're tempted right I'm Val GoPro there's the beta 1.60 1.40 before you

use that down I want to keep going down see a standard of the beta B tell me what the standard error what what the standard error there for the beta

50 right my regression beta is 1.60 my standard error is 0.50 I might as well have told you absolutely nothing right the true beta could be somewhere from 6 to

2.6 that's what I mean by noising nice thing about Bloomberg bait is they give you the bait and they give you the noise term everybody ignores the noise term maybe you need to draw attention to

say why you ask me to use the beta look at the range around the beta now once in a while though you might get a great looking beta you should be even more

terrified or what you just found this is a beta page I've saved for almost 25 years it's a beta page I printed off for

Nokia in 2000 this is amazing regression R squ is close to 100% standard a is close to zero every point is on the line right If This Were A statistics class

I'd get a go medal but I have a question why is my Nokia beta looking so much better than my GoPro beta is it because it's a more

mature company no if I made it know I could have taken Nokia and you know GoPro and run run it similar I I'd have got a much

better looking regression with the choice as well Danny what do you think look at the index right you know what blg does when you ask for a it's

very parochial you ask for the beta for a German company to run it against the Dax an Indian company against the sensex a US company G the S&P

500 I asked for Nokia which is a Finnish company and guess where Bloomberg is going it's going to Helsinki that's a that's a Helsinki I'll

be quite honest I had no idea what was in the in the hex I thought it was a witches Cur I ran this regression and I took a deeper look and I wish I

had not in 2000 I think Nokia owned Finland it was 80% of the index so what is this a regression of Nokia against

Nokia and what do we finding an awful lot of the time the two move together congratulations if you'd find anything else this would be a statistical artifact you should publish in every

paper in the world this is a useless regression in here why why do we say we use betas it's because the marginal investor is Diversified right so I want to take a

look at who the marginal investor in Nokia was in 2000 it was easy to find I took the top 17 investors at the top of the list was bar and they held Nokia as part of their

Global Index Fund if you're barlay you hold Nokia as part of a global Index Fund you don't measure risk against the hex in fact if you want to redo the regression one of the nice things about

Bloomberg because you can change the index what would you change Danny what what did you do replace the hex with or some Global Equity index and

there is one actually on Bloomberg called the msci global it's the most widely used Global index there's a symbol for it I'll send you the symbol you can take any beta page for noia or

for for on Bloomberg and replace the local index with a global index when I do that I get a more sensible beta page but it comes with a consequence what will the standard error look like it's

going to be much bigger you run a beta regression right a big standard error is a feature not a bug because most of the risk in a company in a diversified portfolio is going to be company

specific risk and that's all you get it from the C question you so if you went and put this against a more Global index

the stand error is say 50 like the other example you're saying how like to me it seems like there's not much validity behind that B There is it basically I

told you I was going to dig a grave for a single regression ba and bu because these are what one pass in Time One stock one index There Is No Escape Hatcher there's no way to fix it I've

seen people sit in front of a Bloomberg terminal play games with the page thinking they can make this better and your point is there's no endgame here because if you do it right you're going

to get a big standard ear nothing you're going to do that makes that go away is helping me but there is a way out and we talk about it's a statistical way out and my my solution is to try to take

that statistical way out so for the moment though I think you can see what the problem is when you look at a single regression paina let's try a different company

remember GameStop what a stock it was right course as a as a business what did it do I mean if you are of the right age and you were remember you know you might all

be too young to do it but there was a time when people used to go to malls if you've never been in a mall it's this big place lots of different stores really nothing to do you had no money to spend but you could just wander around

the mall all day long and that's what teenagers would do until they G to the came to the GameStop store and it'd be very crowded because in the GameStop store you could stop and play games on

the computer you didn't have any money to buy a game but the reason they let you play the game is the minute you got to a threshold you came back to the store and bought the game you

wanted so it's a mall store business brick and mor retail whose time is come and gone and so coming into 2020

2021 the perspective was hey this is this isn't we're looking at the end game here and there were a lot of people who decide to sell short on gamestock we talked about you know this at the start

of the class about how Revenge can sometimes drive it and of course it then became the target of all these the this group on Reddit called Wall Street bets that decide to take GameStop and you can

see that spike in the prices so if you looked at the price vola this is a stock that went from $20 to 400 back to 20 is that a risky stock you think don't think

too long clearly goes from 20 to 400 back to 20 you are incredibly terrified so I want to run a regression B to see how terrified I should be you know what I get got as my regression but it's

tough to read there I got a negative beta for the stock so first I want you to ask a question how can an incredibly volatile stock have a negative

beta what what beta measures is not just volatility but how that volatility is correlated with the market right so if you have something that's volatile for a reason that have nothing to do with the

market it'll actually have a low baa biotech companies have low betas even though they're risky companies because much of the risk is company specific the fact that your drug didn't pass the the

third phase of a FD is has nothing to do with the market most of the risk year in fact all of the risk year is so company specific that when you run the regression the

regression what's happening in the stock has nothing to do with the market in fact it's a little inversely related Market because strange people buying and selling the stock for strange

reasons but here's my question if your valueing GameStop at this time do you really want to put a negative beta into your because what will a negative beta

lead you to end up with a cost of equity that's actually lower than your risk free rate you'd be insane to do that right again a regression beta that's

completely leading you off the tracks and finally there is this phenomenon of gaming that happens once you decide regression ba is the way to

go at the start of the class we talked about bias that often you come into a valuation either wanting a high number or a low

number let's say you are valuing in this case Bombardier Bombardier is a Canadian Aerospace company makes these smaller commercial

aircraft and you want to get a beta from Bombardier bombarda is traded on the Toronto Stock Exchange and it's traded as an ADR in the US so what you have in this page is

the regression the beta that I got by regressing the bardier ad against the S&P 500 or Bombardier the local listing against the Toronto

exchange I get two different betas in fact the betas are 1.70 and 1.21 so let's say your bias was that you wanted to find Bombardier cheap you want to get

a high value for Bombardier you sit in front of a Bloomberg you try these two be honest with me which one are you going to show me as the beta for the company given that you have a bias

the low one because it'll give you a lower cost of equity and you claim you'll claim this this is the beta page for the fact that you tried 15 other Pages before you got to this one is

completely missed what I'm trying to say the next time a banker holds up a beta page for a company before he goes on say how many

beta pages did you try till you got this one because think of the number of things you can change in this page you can change the starting point and the ending point you can change from daily

to weekly to monthly you can change the index you're going to be able to move the beta all over the place which by itself is not the issue but it means that whatever bias you have you'll end

up finding a beta that reflects that bias so we've got the whole deep enough let's bu regression bet is they're noisy you should get even more terrified when they're not because the index is

probably misspecified in periods of volatility the volatility is nothing to do with the market Merc when it was targeted in the voox

lawsuits ended up the beta close to zero because every month the one thing that drove the stock price was did you lose that lawsuit or win on it and finally this gaming phenomenon where if you give

yourself enough time and you can change the parameters you can get pretty much any beta you want for the company so with that lead in I'm going to take a step back when you talk about

betas you are in a sense buying into the capital asset price model and when you buy the capital asset pricing model you're bringing 65 years maybe 70 years

of portfolio Theory as baggage with you and there are people out there who don't like any of this stuff some some high-profile Warren Buffett has been famously quoted as saying you know

Charlie Monga too no bet is you know it's it's a it's it's a terrible way of measuring risk bait is are bad I accept that bait is are noisy they flawed but

what I don't accept is what people do as a consequence they say I don't like betas therefore I'm not going to do intrinsic valuation so I'm going to give you if

you don't like betas or you run into people who don't like betas a way out of this box because betas to me are just a means to an end they measure of relative

risk and if you tell me you don't like Bas I'm going to ask you why so I want somebody to play the game you know there might be somebody here who doesn't like

BS you might have worked with people so if somebody doesn't like Bas I'd like you to volunteer yourself because I'll lead you to the process of what your alternative should be for Bas cuz you

probably were skeptical when you heard about vas in your class Diversified investor we run a regression and I'm building a whole thing so anybody want to be the candidate for not liking

Bas no go ahead diction so what what is it about Bas you don't like like you said I I don't believe that I

can use it in in Valu but but I'm trying to dig why can't you use it right so I'll give I'll give you a starting point there are two big assumptions we make with betas one is the notion that the modu investors is Diversified right

that's an assumption the second is that stock prices allow me to capture that okay so which part of those those are the two big ones so which part makes you

more concerned the fact that the module investor Diversified or that I'm using stock prices okay so and the reason being we're doing intrinsic valuation we

said markets make mistakes why would I use a market price based measure of risk if I'm doing intrinsic valuation this would be what you run into over and over again if you go to Omaha Nebraska and

you talk to Value investors they so that's crazy using betas that comes from prices we intrinsic value people okay let me take that to the next step I agree with you it's a market price and

you don't like markets so you got to give something else right if you're in the intrinsic value space rather than looking at volatility in prices what

might you look at you know what's a much more intrinsic number maybe earnings right the one catches you don't get earnings every day so you have to use quar I could compute

a standard deviation your company's earnings just like I do stock price let's say I come up with 36% I can't use the 36% because I have no idea whether it's high or low so I

want to standardize it what's the most logical way to standardize it compute this for every company in the market let's say it's 24% on average if I divide the 36 by the 24 I come up with a

number that looks very much like a beta right 1.50 and now I'm playing your game you want earnings volatility I'm going to give you earnings volatility you might say look I don't think it's earnings it's debt that I'm

worried about when I think about risk that companies might get into trouble so I can compute debt ratios again my objective is to take what whatever you tell me you think about as

risk and convert it into a number that looks like a beta so prices are what bug you the Assumption we're using prices I can fix it but there are people who say it's a

marginal investor assumption that bugs me I'm in a market where investors are not Diversified why are you assuming the marginal investors divers if that is your reason for not liking baas I'm going to take a different

path I can then what are the reasons we use beta is we look at the portion of the variance that you cannot diversify away that's what beta is you've just told me that you think investors are not

Diversified so you're basically arguing that you should be looking at the total standard deviation something we don't do because we assume Diversified investors but that's a number that I can pull from the

market let's say it's 50% again I'm going to look at that average across all stocks let's say 30% 50 divid by 30 gives me a relative standard deviation of 1. 67 it comes

with none of the B baggage of Diversified investor to me beta is not the place where I want to fight your you know this is not the hill I want to die

on if you don't like betas I'm going to come up with a different measure of relative risk that captures what you think about risk I'm going to put it where I put in the beta so in the case of um standard deviation and earnings

I'm going to put in that 1.5 that I got from and and move on because ultimately all I need is a measure of relative

risk so if you don't like portfolio Theory fine throw it all out right but don't tell me you don't adjust for risk because that I don't get give me your

proxy for risk what you think of as risk and I'll convert to something that looks like a beta so if you don't like the Diversified investors you can use relative standard

deviation know so if you're valuing a private company might say owners are not Diversified you're basically using a Rel standard deviation I call it a total beta and I'll take you through the process you can use proxy models where

essentially you say look I'm not going to even Tred to measure risk with a beta but small companies are are risky big companies are safe technology companies

are risky as you make these choices remember this is a flawed measure of risk as well you're replacing one flawed measure of risk with a different one and you'd ask yourself am

I going to be less exposed to Mistakes by doing this than by using the traditional beta there's a reason I stick with the beta I mean I I'm

comfortable with it I know it's all of its flaws it's like being married for how many years have I been married long time 30 my 38th anniversary

is coming hey by this time we know each other right so in a sense bet is like you know spouse you've been with for 38 years you know you you know you get up

in the morning you know all of its problems and they the beta knows all of your problems and you kind of move on so I'm okay with replac not my spouse but

the beta with something you want to use instead because ultimately my end game is I want a hurdle rate I want a cost the capital I'm going to get there by hook or by crook right so you will run

into people who will make it make this their big thing I don't like pis concede that to them so I'm willing to use something other than betas move

on cuz ultimately want to H me other value investors don't value companies and that sounds like a weird thing to do how do value investors find undervalued

stocks they screen they screen for low PE so basically they're doing pricing and this is my problem with value investing is if you don't actually value

companies why is there a value in your description you're just low price screeners right that's what you are and then you wonder where's the payoff to Value investing so

when value investing doesn't work they said this this shows that valuation doesn't work you haven't valued a company how do you know what works or doesn't if you haven't tried in the first

place last time I was in Omaha I didn't go to the meeting I talked to 200 portfolio managers who are people who come to Omaha every year for decades ask some question how many people in this

room have actually valued a company three people out of 200 have actually valued a company they were willing to talk about good management you know would look at

all I said I I accept all of that stuff but you're basically saying I'll buy a stock with a low p ratio and good management that's a screening device it's not valuation it's not

complete because you're not willing and and so you're using this as an excuse for not having to Value companies and one more thing there are a lot of people who will take the cap risk fre rate plus

beta times risk premium and then they was really these are called what are called buildup approaches to a cost of equity I call it makeup approaches because basically when you get a number

that's too low what do you do you start adding numbers small cap premium 4% still not high enough or let's add an liquidity premium another 3% still not high enough a private company premium

another 9% still not and then you get really desperate a company specific risk premum which you completely make up and add on you've really lost control this process at which point my response is why you even doing intrinsic valuation

if you're just going to make up a discount rate you know in the first place yes small cap premiums though we we do

know that small cap compan do use historically no wait wait what's that history based on 1927 through 2024 I updated just last

week the small cap prum for the US was 3.03% smallest de you know the standard ER in that number is right it's about 1.67% so already statistically pushing the margin but I'm

willing to con see that's close enough there a small cap but if you take that 1927 through 2024 time period and you

break it up into 1927 through 1979 and 1980 through 2024 between 1927 and 1979 the small T premium is about

7% between since 1980 there's been no small care premium when did was the farmer frch study which is 1992 which was 13 years still you were close enough

the the further away you get from 1979 the more tenuous this whole thing becomes it's amazing how much stuff in valuation we do with this lagging

history carrying us along so I actually will will send you the small cap spreadsheet that I created where I said you give me a so basically you can pick the starting point and the ending point

for your time period and as you move 1927 1930 1940 1950 you see the premium shrinking and by the time you get to the mid 70s and you start off every time

period since then the small cap premium has been zero or negative and which suggests that there's something fundamentally that has shifted in markets and I think here's what shifted

you know how difficult it was to get information on a small cap company 60 years ago you I'm not saying you were around 60 years ago do this but getting the data there was an information problem there's a liquidity problem

there was a trading problem there were no ETFs that did just or index funds that did small cap stocks today we have Vanguard index funds that have centered around small cap companies that get so I

think there's a reason the small cap premium has disappeared so when people keep using that premium and basing it on historical data I think they're letting

the tail of the first half of the of the Data Drive the entire process so that's a capm plus there are accounting risk measures yes

arees from being small yeah that's a different question that's always been my issue small cap is just a stand in for something else right because there's nothing inherently risky about being small or large I can think of small

businesses that are incredibly safe and large business are incredibly risky we just allowed it to be a proxy something standing in if you don't like price based measures then you got to go with

accounting numbers and I'll be quite honest I'll take market prices over accounting numbers but that's just me right you can look at earnings you can look at balance sheet ratios you can do

the ultimate Z score if you want and but it's more useful in fixed income than in equity as a measure of risk but something like that you can use even qualitative risk models where you

basically assign companies to five buckets the know value line used to be a big investment you know investment newsletter for a long time they used to

classify companies into risk into five groups one 2 3 4 5 I'm okay with that as well I'll just give different costs of equity for each group I amet as I said

completely flexible on how I adjust for risk if you tell me what bothers you about the way you're measuring risk one final point about bait is before we leave before we establish a

different way of thinking about Bas now as I said most people ask them where the betas come from they say they come from regressions you dig a Little Deeper they actually they they come from a service they come from Bloomberg they come from

B You Dig a Little Deeper they say it comes from a regression but I want to dig even deeper ultimately betas don't come from regressions they don't come from Services they come from choices you

make as a company let's go through the three choices and there only three that determine whether you have a high beta an average beta or low beta tell me what you do as a

company if the product or service you provide is discretionary the more discretionary it is to your customers the higher your B dat will be as a company Jacob what do I mean by

discretionary what what am I talking about discretionary in the sense of I can delay buying it I can defer buying it if that's the kind of product or service you provide you have a higher

beta because in Good Times you will do well in bad times people will defer it so the retail business luxury retail is going to have a much higher bet than department stores

department stores will have a higher bet than discount retail and discount retail will have a higher beta than grocery retail because you can

live without your Gucci but you can't live without food but even within grocery retail you'd expect Whole Foods to have a higher beta than Kroger I went

to both stores yesterday the eggplant was three times as expensive at Whole Food but was organic when you doing well you're making a lot of money you go for the

organic eggplant you lose your job trust me you'll be okay okay with the inorganic stuff cuz you're going to die anyway so you can see even within

businesses you can see variant sprouts and Whole Foods will have higher betas than Kroger's or the Walmart Target grocery version where you go for the lowest

prices now they mentioned the company he's working on is on right Sports work I've gone on their website I've never

hit the buy button I just go to look because seems a little expensive to me maybe because I'm old $150

$200 it's Cloud it's got big big heels supposed to make my feet feel better I would expect on to have a higher bet than you know something that

I just put on my feet and the problem in the re the Footwear business is that all brand name companies are all pretty highly priced it's not like you can get you know cheap Footwear but basically Al you're looking for how discretion the

product is the more discretionary the higher the bet that's a first stop so the company you picked and I'm assuming you pick the company I want you to start thinking about what they do and if it's

a b2c company it's easier than a B2B company because b2c you can think about the product would I be able to put delay buying it or defer buying it and the answer is yes I it second stop tell me

something about your cost structure why does cost structure matter the greater the proportion of your cost that are fixed cost cost the higher your beta will be as a company think of what when

you have a lot of fixed costs in Good Times you make lots of money in bad times you lose lots of money in

2020 Marriott collapsed as a company why because Co shut hotels down and its cost structure remained with those huge fixed cost you had to keep those hotels cleaned up you couldn't just let you

know let everybody go because people would move into the hotel rooms and never move out but Airbnb reduced its cost by 80% why because it's a platform

you send people home and you have less costs I would expect Airbnb to have a lower beta than marot it's got less fixed cost so with your company I want to start thinking about cost structure

what kind of cost and I'll give you Clues you can look for in the financial statements that will kind of tell you whether your company's high or low operating leverage and thirdly when you

borrow money you create a fixed cost you did not have until you borrow the money interest expenses in Good Times no problem you pay the interest expenses in bad times you're still forced to pay the

interest expenses which makes your Equity income much more volatile that's it those are the three things that drive your beta what do you do product and service how discretionary is it what

does your cost structure look like and how much have you borrowed now do you see why airlines are insanely structured what's a cost structure look

like for an airline incredibly fixed right the aircraft the fuel the the employees and if you have a lot of fixed cost as a company what

should you do in terms of borrowing money borrow as little as you can you already have a lot of fixed cost what strikes me and this is part of the

insanity of how slow companies are to adjust to changes is Airlines to borrow money on top of the fixed cost and then they wonder why they keep going bankrupt

every five six years right the reason they do it is prior to 1977 when you had before you had deregulation you could do Cost Plus

pricing so effectively control the pricing so you could pass it through that's no longer the case so as you think about Bas I want you to start thinking about your

business economics in your company yeah borrow less right but then why infrastru compan have why so traditionally infrastructure companies have borrowed money to build

infrastructure why did AT&T why was AT&T able to borrow money when it was Marbel in the early part why were they able to borrow money and get away with it what were they building

into a monopoly that was a trade-off legislation was actually passed protecting AT&T from competition why cuz without that being passed there's no way

you spending billions of dollars on infrastructure and not control the pricing and there in lies the so why isn't more infrastructure being built in India I you know a student of this class

became the head of Morgan Stanley infrastructure and he quit after four years and here's the problem he said we're spending a lot of money the government tells us we will have this

nice secure Market but they keep changing their mind and if they keep changing the mind we're not spending billions up front it's a challenge with infrastructure

with privately funded infrastructure is without giving pricing power then you're really are you know it's really difficult to figure out the economics of

how you make this work you have like a long contract the Govern then might help you but you know but it's not just a contract the pricing has to be set as well so if the government says we will

pay you a price that'll be renegotiated 5 years from now having a 50-year contract doesn't quite help you right so it does mean that without that pricing power it becomes much more difficult to

carry that fixed cost so here's my way around regression betas I don't like them I haven't used them in 40 years in valuing a company but I need to come up with something to replace

them here's what I'm going to do I'm going to start with the bait of the business or businesses you're in sounds difficult but it's actually you tell me the three or four or five businesses

you're in I can actually tell you roughly speaking what the bait of being in those businesses in fact if I have the information I might even be able to correct that b for the fact that you

have higher fixed cost or lower fixed cost if you're if I gave you the betas for hotel companies and you're looking at Airbnb you might want to use a lower beta so the Second Step I'm going to see if I have to adjust for operating

leverage and I'll talk about the informational problem you going to run into and the third step I'm going to bring in the debt you have as a company because that too can affect your beta so let's start with the first

step when you want to estimate the beta for a company you first have to figure out the business the company is in and I want you to be careful about defining business because there are two

mistakes people make often they Define business too specifically my company is in the shrimp fishing business no no no no your company is in the food processing business because the minute

you start to Define it as very narrow segments you're going to run into trouble even thinking about the risk of that business I'm going to find publicly traded companies with business and I'm going to clean up for it but when you do

you're going to get what's called a pure P play beta beta for being in this business and I'll talk about the mechanics of doing this so if you're in one business I'll give you that one Pure Play bait if you're in three business

I'll take a weight to average so essentially this key part in this process is finding the business you're in and finding the betas of other companies in that business so that'll give me a beta for

the business a steel company or in the steel business I can give you the beta a pure play beta for being in the steel business but then comes the Stony issue of within the steel steel business

aren't there some companies with higher fixed cost and others with lower fixed cost the answer is absolutely and I'd love to be able to correct for it I know mechanically how to do it but you know

what the biggest challenge in correcting for operating leverages take a look at your company's income statement or financial statements collectively and I challenge you to tell

me how much of the cost in this company of fixed costs and you can make guesses right you can say cost of good sold is probably variable cost sgna is probably

fixed cost but you have no wayno the biggest challenge in correcting for financial for operating Leverage is getting information on fixed costs and

variable cost at individual companies if I can do that I can clean up for the beta so I don't have to assign every Steel company the steel company steel business beta I can

actually but to do that I need to know fixed and variable costs you know how often I I correct for operating leverage almost never because the information is not

there the exceptions of when I have to Value company like Southwest in the airline business Southwest has the most flexible cost structure of any Airline it would be unfair for me to attach the

bait of an airline to Southwest it's flexible cost structure comes from lots of different choices it makes first it doesn't fly for the most spot doesn't fly into the it doesn't fly into hair it

flies into Midway why because the gates cost less so every choice that Southwest is made over its life has been to reduce fix cost the way it pays off is a more

flexible cost structure lower beta so Airbnb in the hospitality business so unless you have a company look and say I need to correct the beta I would just let it go because the getting the

information and fixed and variable cost is really tough to do so for the most part we will assume the operating leverage with within a sector is the same but if you have a company that's

exception try to clean up for that which brings me to financial leverage and here we know how to adjust B in fact in 1971 Chicago Professor

Robert Hamada came up with what's called a Hamada equation what's a Hamada equation basically he said you tell me what the beta of the business you're in you tell me how much you bought with the

debt to equity ratio I can tell you what the beta of your Equity will be now I'm going to make it a little confusing because up till now we've just talked about beta the unlevered beta is the

beta of the business you're in the lever beta is the beta of the equity in that business pause right there can you have a business that's safe low UNL beta and

end up with a high Equity beta yeah if you borrow enough money your 30 is if they were completely Equity funded whe beta is like 0. 2.3 Banks if they were

entirely Equity funded or bet is like 0.25 what makes their the betas go much higher is they use a lot of debt and you can see why they can get away with it if

you have a low UNL beta adding debt doesn't make you kind of topple over you're not like an airline where you're taking a risky business and adding debt on top of it you have a core business

that's safe so that's called the Hamada equation it's a it's an equation that's built into all my spreadsheets but it comes with a potentially fatal flaw it's

built on the assumption that all of the market risk in a company is borne by the investors or put differently it assumes the beta of debt is zero it doesn't it

doesn't assume that debt is riskless it just assumes that the risk in debt is not Market risk just default risk and that's not a bad assumption if you're looking at a single a a double A or

Triple B rated company because what drives those rates has nothing to do with the company it's driven by what's happening to the t-bond rate what's happening in the macro economy but let's say you have a company with Triple C

bonds any of you trade in the high yield bond market high yield bond market it it behaves a lot like Equity what drives the bond is more to do with the company

which is another way of saying there is beta risk or Market risk with that Bond the bait of that debt is not zero so when people make this there I

mean people will try to fight you in every step of the process because they really want to do whatever they want to do this all gets in the way so they don't like the beta you get over that they'll get to this point say but the

beta of De is not zero okay tell me what it is that'll usually shut the conversation down because they just want say bait of debt is not zero I'll help them out I'll give them a bait of debt because if you're Triple C rated I can

figure out what the bait of debt is and if you can there's actually a version of the equation where you can bring in the bait of debt what that effectively does it transfers some of the market risks to

the lenders which means your bait of equity will actually be lowered high debt ratios which actually makes sense would I take the time to do this for most companies it's not worth it triple

be single a it's not but if you have a really high yield bond company Triple C Double C it might make sense to make this correction because you want to kind of parcel out some of the market debt to

the lenders which leads me to what I call bottom of an fancy word sounds like I'm doing something amazing but I'm not so wi is going to start a company and

I'm going to put him pick any two businesses you'd like to be make them fun business don't say steel and chemicals pick some Airlines how about Airlines and

cannabis it's a good business to be right just make sure your Pilots are not smoking before they get on right so Wht is in two businesses Airlines and cannabis and he wants me to estimated

beta for his company so here's what I'm going to do I'll be back tomorrow with I go to my office I find as as many publicly traded Airlines as I can probably

100 and as many publicly traded cannabis stocks as I can are there any out there no there are a lot of them in Canada the reason you don't have

cannabis stocks in the US is because you have a patchwork of laws covering the US and you can't as a company operate in the US if you're going to be arrested as you fly from one state to another right

it's a problem so but there are about 30 or 40 cannabis companies in Canada so publicly traded I get the betas for each

of them then I take the average of the 100 Airline betas and the 100 cannabis stock betas they're regression betas right are regression betas lever betas

or unlevered bers levered because the debt is in the stock prices they reflect the leverage of airline companies with the airline bers and cannabis stocks the Cannabis bers I take out the effect of

the debt using this equation it's called called unlevering a beta but I'm just taking out the effect of De I now have an unlevered beta for being in cannabis and Airlines I come back to you and say wait you told me you were in two

businesses can you tell me how much value you get from each business and your reaction might be I don't know how much value I get but I get 80% of my revenues from Cannabis 20% from

Airlines the only way I can get people to get on my Airlines is by you know making them smoke a lot of weight I take a weighted average of the two unlevered bays point8 times the Cannabis beta

point I now have an unlevered beta for your company ready for the last question I'm going to ask you do you have any debt if you say no I'm done right the UNL bet is L beta if you do have debt

I'm going to convert that debt to a debt to equity ratio say 20% and I'm going to come up with the lever bait of your company why am I doing all of this because I don't like regression betas

right where did I get the 100 Airline betas regression all I've done is replaced one regression beta with an average of 100 I'm going to come back to you remember those regression betas you

run them right your big standard errors how is using an average of 100 regression betas helping me deal with that problem

let's say each of the 100 regression Bas has a standard a of 0.5 they're all terrible looking betas noisy betas what's the a what's the standard

ER of the average of 100 betas going to look like remember the law of large numbers in statistics what the law of large numbers says and I never got this in my

statistics class you take a 100 crappy numbers you add them up and divide by 100 and magically you get this really precise number but it's not

magic here's what's happening a big standard means some of these betas are overestimated right some are underestimated when I average out guess what I'm doing I'm averaging out my

SS an average of 100 regression betas is going to be 10 times more precise than a single regression beta you saying you're making up stuff I do sometimes make up stuff but this one I don't have to make

up how do I get from a sample of 100 to 10 times more precise it's the square root of the sample you're saying what if I have only 49 companies I'm picking nice numbers that you can take the

square root of it's going to be seven times more precise what if you have only 25 companies five times more precise what if I have only nine companies three times more precise anything beats

one so if you're saying I'm using a regression beta because I have only five companies guess what the square root of five is still going to give you a more

precise beta than any single regression beta I have 48,000 companies in my public company Sample When I compute industry average betas there are some

sectors like chemicals where I have 693 companies what AOW allows me to do that is I cut across borders when I do betas I'm not doing just a beta for us companies when

I'm looking at a US company because the average beta in every single Market is scaled around one the nice thing about betas is they travel across borders I can combine companies in different

markets and still get away with it so the big I mean I can give you multiple reasons for bottom of betas but here are the three biggest ones you get less noisy betas you also get more updated betas if

with was just an airline company till last week and he entered the Cannabis business last week there is zero chance at a regression ba will reflect that right but I can pick weights in fact I

can be proactive and say what other businesses do you plan to be in and he might say I'm going to going to the casino business next week and the alcohol business the week after he going

full-time into s the anti ESG stock I can just estimate a beta for him based on the business I regain control of this process remember we talked about stories you tell me a story about your company

doing other stuff I'm going to bring it into my betas through the weights and if I came to you with a private company that's planning to go

public remember the old way you had to run a regression you can't run a regression for something that doesn't have past prices but I can do a bottom up beta for zato at the time that it

goes public even though it's not been traded before cuz all I need to know is what business soato is in and then I can get a beta for any it gives you the freedom the flexibility to Value

divisions in know you know private companies so I'm you know as I said I don't use regression individual regression betas in evaluation I haven't use them in a while because every time I

come to this fork in the road do I want to go with a single regression or a bottom of beta to me the bottom of beta track looks much more attractive so just to give you a couple

of examples a bottom of betas in space this is actually a beta page for Val Val is an iron or Mining Company it's a Brazilian company and there is the breakdown of the four businesses that

Vol is in I pause there how would you know what businesses your companies in where would you find this usually in the financials you're required to you get more than 10% of

your revenues from a different business both IFRS and GAP required to break it down that doesn't mean the breakdown is going to make any sets I've seen companies break their businesses down

and say what the heck is that but in most companies this is where you start V broke themselves down into four businesses so there are the four businesses there's my sample to get the

bottom of a and you can see the size of the sample 693 Global special chemical you know for the so for each business and with each of those I get enough companies in my sample that I feel much

more secure about the unlevered betas that I get so I have the unlevered betas of the four businesses what's the final step I need weights for the four businesses V actually told me how much

they got his revenues and if I were in a hurry I would just use Revenue weights you saying why just not hurry because a dollar in revenues in one business can be worth a lot more than a

dollar in revenues in a different business driven by what usually difference in margins you're in a low margin business and a high margin business the dollar in Revenue in the high margin business should have a

higher value see that sounds like a lot of work I used to show shortcut the same companies from which I get the unlit betas I look at what multiple of revenues companies in that space traded

Enterprise so EV is the Enterprise Value a typical metals and Mining Company trades at about 1.9 time 97 times revenues I multiply your Revenue by that

number I come up with an estimated value so see the second last column those are my estimates of values of four businesses I compute weights on this and

based on my numbers Vol looks like it's about 76% iron or about 177% metals and Mining about 5%

fertilizers and about 2% Logistics just make sure the weights add up to 100% take a weighted average I've got an UNL beta for Val as a company in my corporate finance class

here's where we'll diverge because there it's about managing businesses if you the manager at valy the CEO the CFO the last thing you should be doing is using the same cost of equity for these different businesses right they very

different risk I can now tell you what the cost of equity should be for each business because I have a beta for each business that I can use to build up to a cost of equity so not only does it allow me to

manage multi- bus companies it also allows me to value a division of a company if Val I think right after I did it sold its Logistics business if you're going to sell the logistics business the

beta you would use to value that divesture would be the beta of the log istics business not the bait of the company because ultimately you got it goes back to the to what we talked about

with the kenot case right you tell me what you're valuing the target company's discount rate that should reflect the risk of the cash flows that was Vol let's look at imra

Brazilian Aerospace company so when I first look for peer groups I start narrow so I said it's Brazilian Aerospace company let me go find other public publicly traded Brazilian

aerospace companies and I ran into a bit of a prom you know what my sample size looked like one it was embri so replacing embri with EMB so I said let me look at Latin American aerospace

companies guess what my sample size still was one still EMB and then I said why am I staying focused in Latin America it's an aerospace company it

makes aircraft to sell to Airlines why can't I compare it to Boeing and you know Airbus and that's effectively what I did I brought in global aerospace

companies and the UNL beta that you saw for Global Aerospace was 0.95 and I do this routinely but there are times when it might not make sense in other words using Global

averages when I do this what am I assuming that your underlying business risk is the same no matter where in the world you are and for Aerospace I think that makes sense it's driven by Airline

demand which has nothing to do with you know if if you're a mining company I think it makes complete Sense To Go Global because what does it matter where you are ultimately you sell but the

times when it will not make sense of what I want you to think about Julian structur you know how does a let's take any aircraft company right the fundamental business

model is the same you got to spend huge amounts building this infrastructure to make aircraft often with 78 n years of leading before a new aircraft comes in

that they share in common all all three now there are company specific differences right because you might say look you know the fixed costs are higher in the US than they are in Brazil I would love to adjust for that but

fundamentally in Aerospace the model has stayed the same pretty much for much of Aerospace history the same thing was true for automobiles until Tesla kind of broke open that mod so disruption can

sometimes change it but Aerospace has been kind of immune from that bomb bardier you look at Bombardier embri you know Airbus and Boeing they all have very similar structures as to how new

aircraft roll out and what the lead time and lag times are so I'm not sure that there if there was enough of a difference I might start to kind of look at some other sector that be a better so

if you feel the sector doesn't make sense look for something else that does well when you do well and does badly when you do badly right because that's ultimately what you're looking for now

I'll tell you when you might not I'm on my web page I do report Global averages for pretty much every Industry Group and for a lot of companies I just go with the global averages but I also report US

averages Europe averages Emerging Market averages and I do Japan and I throw in Australian Canada because I'm not sure where to even put them in to be quite honest I me they don't quite fit

anywhere else you know when I end up using Emerging Market averages is you have a business which is discretionary in a in in an Emerging Market but nondiscretionary in the US

and Western Europe used to be true for Telecom know the early days of Telecom everybody in the US already had a phone I'm not talking before smartphones so you could use the Telecom

beta for you but in much of the rest of the world a phone was a luxury you would get to it if you so to the extent that these countries did well and grew you

might get more sales so if you're looking for the beta for a telecom company in those days for an Emerging Market company it might make sense to focus just on Emerging Market telecom

companies banking is the other place where I'm a little reluctant to go with global averages why because the regulatory overlay varies across the world so there I might say look I want

to look at the beta for African Banks if I'm looking at a Nigerian bank right but you want you you're walking this very fine line right because if you try to be

too narrow in your criteria you're going to end up with too few firms you want to get enough firms so you get the law of large numbers working in your

favor one final Point uh in much of what I do almost all of my valuations I look at gross debt what's gross debt is the total debt that you see on the liability side of the balance sheet two billion

three billion 5 billion in much of the rest of the world though people talk in terms of net debt what's net debt you take the gross debt and you net cash out so in fact much of Europe when people

talk about debt ratios they're talking about net debt ratios The netting cash out and often you get into the issue of is it okay to use net debt ratios of gross debt ratio the answer is

absolutely as long as you stay consistent all the way through let's say using the gross debt ratio which was 19% I came up with the beta

1.07 if IID used their net debt ratio I would have actually end up with a negative number you saying how is that possible because a cash balance was greater than the grow

step don't do things like you know I can't use a negative number round it down to zero you have to stick with a negative net debt ratio what that'll give you is an unlevered beta that's

actually lower than I'm sorry lever beta that's actually lower than your unlever beta if I just pause there with the beta .93 I'm

going to end up with a much lower cost of equity than a beta 1.07 right so you saying isn't that better to have a lower cost of equity but remember your cost of equity goes into cost of capital

calculation if I'm doing gross here's what's going to happen my Beta will give me a cost of equity it go to into a cost of capital where the gross debt ratio will be 15% or 14% based on the debt to

equity ratio which will then give me cost of capital that's a weighted average remember debt is cheaper than Equity it'll pull my cost of capital down but if I do net debt ratios and I get a cost of equity based on the net

debt ratio and I compute a cost of capital what's the weight on My Equity going to be more than 100% you got to hang in there your debt ratio will

actually be negative leave it as is the ending cost of capital you have will actually be very similar in the both both the approaches but only because you stayed consistent so if you feel more

comfortable using net debt ratios I I'm absolutely okay with that just stay consistent with that choice all the way through so let's summarize all of the different inputs in cost of equity

because I'm we're going to put it to sleep and start and cost of debt and capital in the next session risk free rate Choice driven entirely by currency nothing to do with incorporation your

Equity risk premium I prefer an implied premium but if you want to stick with historical premiums be willing to defend it make your implicit assumptions explicit and say this is why I'm going to do in a

beta don't use a regression beta try to do a bottomup beta and that might require that you might have to do some gymnastics in terms of determining what business of businesses you're in but I

actually find it less work to estimate a bottom up beta than a ression beta because that's why I do the beta I do it for me so in I value steel company in the middle of the year I don't start from scratch I go look up the beta for

steel companies that's what I use because the betas for businesses should not change over a week a month or even a year it's basically a beta that should be pretty

stable so I will leave you that and see you on Monday thank you I'll be coming