In Pursuit of the Perfect Portfolio: Myron S. Scholes

[Steve Foerster] Hello, I'm Steve Foerster. Welcome to our project "In Pursuit of the Perfect Portfolio. Today I have the pleasure and honor of speaking with Professor Myron Scholes famous for the Black-Scholes option pricing model. Welcome Myron. [Myron Scholes] Thank you. [Steve Foerster] Our project is is focusing on an elusive concept perhaps the perfect portfolio and and we'd love to hear your insights in terms of what is the perfect portfolio mean to you. [Myron Scholes] Well obviously that perfect portfolio does not necessarily exist but the idea of the way I think about the perfect portfolio is

to really think about what investors are interested in. Investors are interested in my view on terminal wealth, they're interested in compound return, and they're interested in drawdown so they would like to for a level of drawdown have the best experience they possibly can. Now the problem with a lot of what we have done in finance over the last number of years is forgotten about compound return, we forgot about growth in our portfolio and we forgot about drawdown. What we have done is talked about relative return we've talked about relative how we are doing relative to

a benchmark so we doing better than the standard Poor 500 are we doing better than a 60/40 strategy in terms of compounding return and getting terminal wealth enhanced through our activities. The interesting point about average returns, or Sharpe ratios, or information ratios which look at the difference between what we're doing and the benchmark portfolio is really ignoring the most important part of investment and that's the absolute return. And the interesting part about investing that is being ignored so moving to what I think the ideal portfolio should be which is concentrating on absolute return and not

relative return is that relative return ignores the benchmark itself ignores the risk of the benchmark itself. So the analogy I like to give is that who's someone on the deck of the Titanic and the music is playing and everyone is very happy to be on the deck of the Titanic because those in the lower deck are dying first and basically when you have the idea of benchmarks or average returns you're ignoring really the most important part of investing which is compound return the average return is a flawed measure it might be able to evaluate whether

a manager is outperforming on average but it doesn't talk about the ideal portfolio and investing. In other words the problem is when someone is thinking about crossing a river you don't tell the person if they can't swim then on average this river is only a half a foot deep because the individual is very interested where they're crossing the river now. Right, that's the most important thing and you're gonna ask them more questions how deep is it where I'm crossing or and because you only have one run of time you don't have an ability to have

average you can't and it returns are not averages. When we start off an investment we start with $100 today if that hundred dollars falls by 50% today or tomorrow and the investment and then triples thereafter it'll go from 100 to then it'll go to two half so 50 and then triples to go to 150. Or if you invest today and it goes to triples first goes to 300 and then goes to 150 by having at that time it will still be 150. So the interesting thing about investment is not a horizon of five years or

ten years but the fundamental question is that what happens each period of time? So the ideal portfolio must not assume that I have a 10 year horizon it must assume what happens each period of time in the compounding process because compound returns multiply they don't average. This when you start off with a hundred you go to 120 you're now investing 100 you initially invest it and the 20 additional dollars if you then lose 20 on top again 20 percent you lose 20 percent of your original hundred and 20 percent the 20 that you left invested

that ends up at 96. And similarly, if you go to 80 to start with you're investing only 80 and it makes 20% it goes back to 96 because why you didn't invest an additional 20 when you talk about average return we're assuming always investing 100 and in in the strategy so the interesting part about compound returns if we refocus our attention on compound returns and what affects compound returns that's the key another key of this problem of the ideal portfolio is we've assumed normal distributions we've assumed constant volatility and constant mean of a portfolio and

that and we know that's impossible you can't have an S&P 500 portfolio and assume that the returns are constant the interesting part about investing is if cross-sectional diversification is free that's claimed by academics and others diversify cross-sectionally if that's free then it has to be the time diversification is also free in the sense that and you can prove mathematically this is the case that if you have a target level of risk for a given levels of return okay then basically if you allow your risk to fluctuate around the target then it's the case that you've

taken excess volatility or idiosyncratic volatility over time and I'd like to refocus our attention away from the cross section which is everyone concentrates on and think about time diversification or the effects of time on portfolio performance to do that we have to ask the question what's the most important thing in terms of time when you only have one run of time you only can cross to cross the river many many times on average okay you're fine but if you cross the river at the 20 foot part of the river and you can't swim you draft

you don't get back again one leg to live one life to live and so in return space the most important thing in returns is the tails of the distribution of returns the idea that you could lose a lot of money or you can miss an opportunity to make a lot of money in your performance of your portfolio and how that compounds over time and so I think the distribution returns are changing all the time if you have an index fund such as the S&P 500 there's no way the risk of the S&P 500 can be

constant over time the composition is changing sometimes technology has a larger weight sometimes utility companies have a larger weight so the volatility or risk has to be changing of that index that's number one number two is the fact that at times the correlation structure among the assets within the index changes sometimes there's a more idiosyncratic list sometimes most of the risk is done by all the assets moving together either down as they did in 2007-2008 or up as they've done in 2012 or so you know when the market went up a lot and so the

core the diversification is not there times when the correlation structure changes and so thinking about the assumption of constant correlation thinking about the assumptions of having constant means and constant returns are fine from a theoretical point of view but it's a one period model it's not a multi period model you are always associated with the the well-known black-scholes option pricing model you've had lots of questions about it perhaps you can talk a little bit about how it came to be and also if you could comment on there are a wide range of views of not

only options but derivatives in general some argue that they're weapons of mass destruction and and perhaps you could talk about those two themes okay Steve first of all you know there's two aspects of the option pricing technology and model one is the technology itself and the other is the model and the technology was at Fischer black and I and Bob Merton developed was really trying to think about how to create a replicating portfolio that's by a combination if we have a stock a combination of stock and bonds that would replicate the returns on the option

now the technology allowed us to have every period of time a changing risk or changing volatility and the changing interest rate and to be able them to think about how that hedging portfolio could be established each period of time and how it would evolve over time now what we developed was a differential equation which described how the option changed with regard to changes in the time and and and the interest rate and volatility and you know the expected return fell away because we had a hedging portfolio or replicating portfolio so I didn't care about what

the expected return was for the underlying security for that period of time and the interesting but we did care about this volatility we did care about that and we did care a lot about the interest rate and time and so that problem was that Fisher and I initially and took a very long time to try to think about how to solve this general case so what we did was we assumed you know that the interest rate was constant and we assumed that the volatility Wisconsin we got a nice ball even though we knew that was

false right we got our model which gotta start somewhere I guess well which took low-hanging fruit we couldn't do it a general case without you know numerically trying to do that we had to do that in a in a way that would be numerical and if you wouldn't understand it as well so what we did is we assumed that the interest rate was constant and we assumed that the volatility was constant and then we ended up with this call the black Scholes model now Fischer black and I used you know the tools we have and

assume that you know that investors could set up the replicating portfolio over a very short period of time allowed time to compress Bob Merton used his technology which is the ito processes and the like and ended up with a differential equation Fischer black and I and Bob Merton you know had lots of discussions about what was the correct approach or what we approach was more susceptible to really evaluate the options but and basically it was more of a theoretical distinction but we always liked the our approach better than Bob Merton this approach but we obviously

respected his approach and it thought it was really a terrific approach as well so I think that the derivation of our the black Scholes technology and model allowed one to you options and initially we have had a lot of empirical validation over the years as to the import of the option prices that we have seen has giving us information about risks as I said earlier in the marketplace I think that the whole development or use of derivative technology allowed for us to change the whole nature of finance what it has allowed us to do is

go from the big okay to more individualized more it isn't kradic more things that a particular entity a corporation needs or an individual needs and away from when I started the profession and Fisher black and Bob Merton were in the profession as these things were big and what finance does and what we've had in terms of financial innovation is the ability to actually compress time make things faster do things more than individuals want and to is to make them more individualized right and three is to make things flexible flexibility is optionality or the idea to

be flexible all three of these things I talked about speed of doing things individualization and flexibility all involve options they've had huge valuation and derivatives have huge implications for our society and ability to innovate and create more things that entities and individuals want in terms of how they manage risks or how they actually transfer risks and also how they build new instruments and securities so it's not as if derivatives replace other instruments it's not as if derivatives could be a complement to or a substitute to various other of how the economy operates but fundamentally they

help you do things more quickly to help you do things more individualized and they help with flexibility and how one manages their portfolios and how would you respond to to claims that they're that they could be used as weapons of mass destruction well the interesting part is that even if I remember correctly if it was Warren Buffett who coined the phrase weapons of mass destruction I think what he was referring to was at the time he acquired general reinsurance there were many many long-dated option contracts in the portfolio twenty years thirty year contracts and that

when he bought the company he realized that the liability for was much larger than he had thought when he had actually acquired the company because the payoffs those options were or the value of the payoffs and they often were much larger than he had anticipated I think that that's what led him to say these longer-dated options were weapons of mass destruction but I do believe that the statement that options are weapons of mass destruction has to do with the ability to lever options or use leverage in options and we also have myriad other ways to

use options or derivatives for leverage or other things other ways in the economy but they do have that levered component now you know again it's sort of survival of the fittest one of the interesting things about a derivative or an option there's one buyer and one seller you know I mean it's a zero-sum game in that sense and it's not and so if I have a buyer and the buyer overpay for the option a seller is willing to come in and write that option and basically you know protect the person in the pricing sense against

against a mispricing or tremendous risk pricing and I think that's forgotten a lot about this when market prices fall and derivatives fall and value then other instruments also fall in value that I think that the fundamental question is are the prices the best or accurate in the sense of the best estimate and as the market really get out of hand and I think no that's not been true you don't see that over time the market pricing of options conveys much more information than does the spot markets you had that in 87 crash you know the

futures market had much better pricing than the smog Marya the spot market wasn't even trading it was completely asynchronous well the option markets on the portfolio's were giving much richer information to what was happening in the marketplace it's true that some people will lose money some people will make money in options if they misuse them just the same way as people who put all their money into a valiant drug stock or whatever and it collapses in value lose money as well I think that the reason options or derivatives have had a misnomer or Mis named

is simply because they're the newest ones on the block you know the same way as if we had had electric cars or self-driving cars to start with you know given the technology that is being developed and will be developed and we wouldn't allow humans to drive but the only reason why humans are driving cars and causing the self-driving cars to have difficulty because humans don't drive as well as the self-driving cars can drive at the current moment so I I think that you know that there's it's always when people want to find something to blame

they tend to blame those things that are new and so there's a tyranny of this at its core there's a tyranny of the herd what exists first but if you look at the extent of which derivatives are still involved in me and have even grown dramatically since the 2007 2008 crisis then wanted to be amazed to say if these are such awful things why are they still being used so dramatically it's you know Stiglitz Stiegler once said that you know survivorship is a very good method of determining value and they survive and they flourish and

they grow now it's true that certain things in the crisis of those 7:08 came to the fore namely that AIG had this price contracts because that but that was an internal control problem with in AIG it wasn't something and it's not the derivative themselves you know people want to write derivatives even if they're a fairly priced if you write derivatives to the extent you can lose all your money by doing it fine but you're making you're making a little bit you know one of the interesting things about writing these options even on Triple A structures

is that you're gonna make a little money a lot of the time but occasionally you take a big loss there's nothing guarantee that you're gonna do that if you don't so it's the risk management issue within the firm a governance issue that counts more so than using these instruments and so the way I think about the market the market gives us tremendous amounts of information about how risks are changing in the market and one of the interesting parts about risk changing in the market is that the option market among other markets but the option market

tells us a tremendous amount about the distribution of future returns the distribution of future returns when we look at a stock the information in the stock price is rich it has but it has two components to it it has changes in risk and expectations of changes in risk it also has expectation of growth or cash flows if it has two things you have one number it's hard to separate the mean effect from the uncertainty effects while the option mark and the beauty of the black-scholes technology and Merton follow-on is essentially it decomposes it and tells

you what the risk is okay because we have proved that you can value an option based on the idea of assuming that it's the risk-free rates the appreciation rate so we have a huge market and telling us what tail risks are which are the most important and it tells you about the entire distribution of possible returns now the mark why is the market options market so valuable to give us information about risk simply because it's the tails of the distribution that are so hard to measure yet those in the option market who are valuing how

the money put options or other than money call options are giving us tremendous amounts of insight as to risk as to the future risk that the market the risk that we see now you say well this the option market is not that far sighted it's only has three-month options or six-month options or your option doesn't have a 5-year option but a 5-year option is not important if you go back to compound return what's the most important thing is what's gonna happen in the next three months that's what can happen five years from now and that's

where the option market has a huge rip richness a huge richness as far as telling us information about how the distribution of returns is changing and so using these information to construct the ideal portfolio one can change the composition of the portfolio based on risk and how risk is changing if one keep the risk of their portfolio cause then you reduce a huge amount of the convexity costs that occur because you allow your portfolio to fluctuate and risk why is that the case let me give you an illustration let's say I call it time diversification

but if you think about a portfolio let's say you have a 50/50 strategy you want to have 50% of your money in bonds 50% of your money in stock let's assume incorrectly that the risk of stocks is the same and the risk of bonds is zero okay okay now 50/50 strategy if you have a 50/50 strategy let's say you always keep your risk 50% stock and 50% bond let's have an alternative strategy the alternate strategy I call that's called the bangbang strategy half the time you're fully invested in stocks the half the time you're fully

invested in bonds okay now if you have no skill that's allowing your strategy than bagging the risk to change dramatic you have no skill your expected return is exactly the same in the bangbang strategy as it is in the 50-50 strategy using the analogy of beta you know if half the time you have a beta one and half the time you have a beta 0 then basically you're gonna have an expected return of one half the bait of that one half the market return if you have no skills so your expected return doesn't change but

your what about your volatility of your portfolio your volatility in the bangbang strategy will be about 0.7 one of the volatility of market because you 100% of the 50% time you're 100% invested in stop while the portfolio of the 50/50 strategy always having a vaild of 0.5 or 1/2 the risk okay it will give you 1/2 the volatility of the market so time diversification is volatility management because it affects your convexity cost and if you reduce your convexity cost that's free that is free diversification but we have ignored in time diversification in the way we

think about investment and I think the ideal portfolio has to involve a lot of discussion about time diversification and thinking about how to obtain information about how risks are changing adjusting the portfolio take account of time diversification interesting and not enough in finance and investment management we see broadly most investment management have said I want to be measured relative to a benchmark I want to be measured relatively I don't want to be measured absolutely I want to be measured relatively and the reason is because they give up the responsibility of asset allocation to whom this

investor or to the institutional man the institution or to the pension fund or whomever else it is they give up that ability they want to be compensated and how well they do relative to the benchmark or not absolutely so the responsibility of Investment Management is not theirs they're only a component or a provider of service to the portfolio and they ignore changes and risk they ignore the asset allocation problem so the asset allocation problem is the most important and I don't care about diversification I don't care about cross section over it I think that's a

smaller component of how your wealth is gonna cumulate over time so if I can just jump in here so if I'm an investor I care about my terminal wealth or my wealth at retirement and you don't care about volatility you care about drawdown okay I care about the downside and then the drawdown you talked about getting information from derivatives that will give me some information should I be should I be using derivatives as an investor as part of this perfect portfolio to help manage the risk for example yeah I mean if I were if I

was doing and I certainly would do that I mean if you're holding risk in your portfolio they want to have the lowest cost source changing your risk composition or managing your reason and the very liquid instruments that exist in the data management debate your risk management is is really the least expensive way through the derivative market whether it's the futures markets or whether it's the options market ways in which you can adjust your portfolio to manage this risk now I'm not saying necessarily the individual could do that but professional management certainly can do that and

do it in a very efficient way so rely on a professional manager once they understand what what my concerns are in terms of drawdown then then we can use these the interesting the interesting thing is the optimal structure of investment has a lot of agrees of freedom to it mm-hmm one is that we have in my view drawdown is important and why is drawdown important because if you can reduce the drop down then and basically one could achieve a higher terminal value for their portfolio and it's really the tales that have the best the most

important effect so what might my portfolio look like so I'm concerned about the tail risk what what might my portfolio look like to help me achieve my goals well the interesting thing is every investor has to ask in a global sense what the asset constraints are you know am i limiting myself to invest in a more limited set of assets that's the first question and the second car and then once one has that constraint of some form or other than the portfolio can be formed optimally to manage the risk over time and to optimize the

portfolio over time using either either active portfolios or combination of active portfolios and passive investments and then that optimal portfolio then the investor would have to determine the level of risk they want to run a portfolio to be run at so let's talk active versus passive and maybe we could go back in and talk about some of your earlier contributions in terms of passive investing what what what role did you play early on in your career in terms of what now is a multi trillion dollar industry right in nineteen 1968 when I was leaving the

universe Chicago as a newly minted PhD and on my way to MIT as an assistant professor at the Sloan School of Management I spent three weeks or so in San Francisco evaluating the investment management process of Wells Fargo Bank under Wells Fargo back was the investment management area was run by Bill Burton and then John McLeod Mack McLeod was running the management science group so I actually looked at what the management science group was doing in terms of asset management and I wrote a report afterwards saying that they had little skills in the inputs to

their in their management their management process and they had a few clients I recommended that instead of concentrating on active management that they should concentrate on passive management and six months later John McCone called me up and said he would like to and be engaged in research on this idea of passive management because no one had ever talked about passive management before and I want to distinguish between passive management and index fund management index fund management says no tracking error to an index I never thought that you'd want exactly tracking index because of the cost

of maintaining a no tracking error portfolio so what did passive mean to use and passive man to me was just thinking about at the time replica or been close to replicating and index but as the index composition changed we trade off the basis cost associated with having not a perfectly correlated structure with the transaction cost of having to make the adjustment instantaneous layer to make the adjustments more slowly over time and so I'm on Macleod Macklin column phoned me up and said I'd like to see research on this idea I said that I had being

an assistant professor at MIT and responsible for my teaching and research that I had this fellow Fischer black who was a consultant in the Boston area we had several conversations together and he was thinking of setting up his own firm maybe he could be the one who travels back and forth between Boston and San Francisco and I would be with him there and I did describe to McLeod that Fischer and I had been doing research with Mike Jensen on testing the capital asset pricing model and that and so we started an association wells fargo wanted

to have a lot of research done and seeing what we can do passively not actively picking stocks but passive investment management and it led to several papers that I wrote at the time not being an academic was a lot of fun because there's a lot of research empirically and at the same time there are the practical implications of what we were doing and developing a portfolio at that time so in you you are a rare animal in in the academic world having made major contributions both on the theory side and the empiric side can you

talk about the the marriage of the two and the importance of that because it often is is overlooked in our profession well I I think that interesting enough I think that one of the things all of science is trying to do when all of business is trying to do is to see how we can have theory on the one hand and experience on the other hand and bring experience and Theory closer and closer together because we always think you need Theory first okay then you got experience second and my view has always been in Fisher

black and other people's views have always been that they're really rich together because if you can have great empirical testing or experience okay that helps your theory and vice versa without theory experiences meaningless and without experience theory is meaningless right because I had everything in science is inductive I don't care what we say we don't go from first principles because you have to data mining things your inductive and hafta with if your inductive you have to be very careful that you don't gather the wrong data and do the wrong things without some theoretical underpinning whether

it's an economics or other sciences you can come completely nonsensical results and you can add and so I think that at the time when I was starting off it would became obvious to me that we would have to combine empirical work with theory and I enjoyed both of those and think it's very important to our science very important and a lot of ways to do the empirical work that I did there was no data you know we had to develop the data and and I worked very hard to make sure that we did develop the

data and and then as we develop the data we made that data there to the community at large and the community at large was then able to do empirical reserve which then fed back on a theory the theory became richer and the two of them together were hand and glove and you know and some things were rejected some new things were born puzzles came about and into the profession and as a result of that it builds a much richer science and I think the interesting thing is what we're trying to do and in academics is

shorten the time shorten the time from theory to experience and those drug trial drug trials would that be sort of the analogy do you think well yeah I mean even in in a lot of what we do is there's research and development I always think research and development arm is named because it's not really research and developments research and testing its development and testing and they all feedback with each other so if you go into drug analogies you're doing the trial it's testing testing everything is not R&D R&D is the wrong name it's not research

and developments research and testing it's development and testing and then back and forth and that creates a richness now if true yeah I think very careful when you're doing that that you don't end up in the situation where you have a dead end because you you've data mined you know and you've garnered from the past information which then tells you that the future but one of the nice things we have in finance and academics and and financial economics is we have theory and we have a richness of theory we have empirical testing of the theory

and then mapping back into the theory and also a willingness to throw out you know what we think is not working and add to things we think is working and look a lot of the innovation inventions and innovations that I've seen over in my myriad years in that profession came from a theoretical side you know and then they were developed and applied you know passive investment is this said is is what's been applied generally that's about 40 or so percent of the market now is managed passively when when I started and brought to Wells Fargo

we talked to institutions about it you know people looked at us like we're crazy how can you manage your portfolio believing in market pricing see the interesting thing is that why would market pricing work market the ideal portfolio we have to use market prices we have to use option prices the information in these option prices is information you can either believe the market gives you information or you can say the market gives you no information if the market gives you information you use that that's the way I started off I say the market has information

you know the market let's use the information in my faith many people don't believe the market has information you look at the government Federal Reserve policy or other people they say Oh market doesn't have any information we gotta use that and anything we do that's crazy yeah to me it's cuz why not get people are putting their money on the table you know I mean even I think we've had these prices of I don't know if it's legal in Canada I know it's somewhat now illegal in the United States but you can have these markets

you say who's gonna win the election you know they have election mark it's right people say how can a market know anything about elections this comes up once every every four years or so you know and they're having these elections the market is amazing how accurate is relative abundance they get polls do you know they do all this stuff you go to a pocket tells you what the odds are to me that's a huge amount of information and that's we thought it we have to use this information the prices give us information and we can

form portfolios we can know what's going on so if the ideal portfolio doesn't use information in the market to do it it's not an ideal portfolio and so you need to look at the prices and how the market is telling us information so derivative markets are telling us information the spot markets are telling us information the forward markets in other ways they're telling us huge amounts information and I think it's better to use the consensus or the wisdom of crowds you know millions of people making decisions and that that's we're trying to do like we're

trying to think of now the whole world is saying the government in releases a report every court you know there's also the null cast I mean everyday you're getting now Google is doing searches and knowing how many people are buying this or asking questions about that you know so it's we're trying to figure out how to price how to get information and use those prices so what happens when you mentioned 2008-2009 the so called financial crisis where 2007-2008 doesn't send 2008 where where we had some major changes in terms of equity prices going down from

from an individual investor perspective what do you think were some of the lessons that they should have taken away from that and and how did derivatives play into the whole you know I mean I remember that Federal Reserve or other bankers saying it was a bolt out of the blue nothing was there if you looked at the option market though the put options were for forecasting crisis ahead they change in the distributional shape you saw at a financial the financial firms the tail risks the put option prices were increasing dramatically as you went into those

seven it was like a tsunami was coming in a market new to tsunami was coming the Fed Reserve I don't carry they want to they want to look at the data fine they don't want to look at the information in prices fine they can ignore that but the market would wasn't stupid you know market was already pricing this in you got oh is spreads were increasing the options implied accreditor the idea of the LIBOR spreads were increasing you know they Ted spreads were increasing I mean credit spreads were increasing I mean it wasn't as if

this is all of a sudden you know my god this is our affair so yes there's tremendous information in prices and the options market or a derivative markets give you this information you don't have to use it but when the reinsurance premiums go up you know if you have reinsurance premiums that go up and does that mean that risk of insurance is going up certainly it is you know so if risk is going up then you could use that information or not you know and so the market was had tremendous and information in fact if

you looked at it the market all the sectors of the S&P 500 that's 10 sectors they were in the S&P column had elevated tail risks where I'm being elevated and a nice part about it is even though crises if you look historically don't happen that frequently you know millions of people are betting on crisis and the elections every day and millions of people are betting on or you know changing their views and protecting themselves you know it's Darwinian survival of the fittest if you're gonna be out there writing options and the tales of distribution and

you're gonna be wiped out pretty soon in a leopard market unless you have some skills we've talked about many of your contributions to the profession in terms of in terms of obviously the option pricing model and early tests of the capital asset pricing model you also did a lot of work in in the area of taxes how important is that from an investor's perspective and and what are some of the insights that you could provide in that area do we do we overlook the impact of taxes migrators thing to me should if I were to

find assigning a tax policy when I talk about risk management is trying to keep your risk of your portfolio constant or at your target and then not allow it to fluctuate around that I wish it were the case that we would allow investors to adjust the risk to their portfolio as opposed to being locked in through a tax policy that currently penalized is you if you adjust your portfolio only for the sake of trying to manage its risk now the interesting part about tax management are the things I've written on in taxes it's tax minimization

is not the correct model what the tax is it always the fact is that you have there is a cost to paying taxes but there's a cost not to paying taxes as well so if you there's that the implicit return or the loss return that you would have by not in your portfolio and obviously that if you had a time sequence in the Optima or ideal portfolio then why do you want to do is you here managing your risk okay it's much easier to tax manage your portfolio if you're trying to just change beta and

control for downside risk than it is if you have a view of a particular security if you have a view of a particular security or locked into that asset and therefore the tax costs might be higher because you really love that you know on you and you can't manage your taxes efficiently but I've always felt that it's possible that if you're talking about beta risk or managing the risk of your optimal portfolio that it's much easier to manage your taxes within that confined that it is if you have a situation where you have a particular

asset you love or a particular asset you don't love so it maybe if we could come full circle to to the the perfect portfolio so some of the themes you've talked about I should be concerned with absolute returns as opposed to relative returns because I if I only consider relative returns I'm doing less worse than others perhaps in a down market I should be listening to the derivatives market which has some important information what about what my portfolio might look like for a typical investor I don't think that necessarily a buy-and-hold portfolio or an ass

allocation such as a 60/40 allocation is the optimal allocation because the risk of a say an index fund is changing all the time and so I think the investor has to take account of that in deciding on compound return as I said earlier a 60/40 strategy is an interesting strategy which is a common strategy you 60% in stocks 40% in bond I don't exactly know where that came from by the way I sort of maybe historically someone said oh this gives me approximately the volatility that I want you know on average but average volatility is

not as important as volatility each period if you have average volatility 60/40 on average you could have that volatility but if you can manage the interim volatility and keep it constant it's much better a 60/40 strategy or an asset allocation strategy just determined by market weights does not take account of risk at all nor does it take a kind of a return okay and so I was saying even if you assume that the returns are constant we know that compound return is affected by the volatility and by the skewness of the distribution you know that

is we've known that from option pricing technology and option pricing theory but somehow it doesn't come in to any discussion about how to run a portfolio the option theory taught us a lot about how to run a portfolio but somehow has never got into the literature or people have talked about compound return never talked about time diversification as I've described here so the idea portfolio has to start talking about time because we only have one run of time and time is very important and I want us to refocus on thinking about time and how you

run your portfolio depends on how your risk is and how you want to manage your risk over time I think there's three ways of making money in the markets and if it's the case that I'll just concentrate on time the reciprocation but if as I said cross-sectional diversification is not putting here all your eggs in one basket is a good model not assuming that risk is constant of your portfolio or a 60-40 strategy or whatever strategy you have will be optimal each period of time it's also something that is free if you want to readjust

to reduce the convexity cost or the compound return drag that you have by taking excess volatility for the average investor can you explain what you mean by convexity risk well convexity risk is the idea that basically if you had a situation let me give you the illustration if you have a choice and let's say your portfolio would have fluctuation plus 20 minus 20 that if on average is 0 that plus 20 minus 20 is not a very good result because if you make 20 percent you know and then lose 20 percent you're down at 96

for a $100 investment if it goes to 80 only recover back to 96 so the convexity cost is 4 percent in this case and we know that the greater than volatility if you have a greater volatility portfolio you're gonna have that convexity cost because the greater the volatility the more you have the loss return because the volatility effect just that bounces back and forth that volatility hurt you in terms of compound return on the on the other hand if you have a strategy which has a target level of volatility and you allow your volatility to

bounce around that target level of volatility that's wasted convexity costs in other words it's it has if you really have a target volatility of twenty five ten right and you allow it to be minus twenty thirty thirty zero or some so on average it could be ten we've been awaiting that reduces the compound return of your portfolio because you've taken excess volatility the more xs/small to you take the more you have lost compound return and so if you don't ever manage the risk of your portfolio to keep it at your target you have excess volatility

that excess volatility has a huge cost let me give an illustration I mean let's say you had a KO loss okay you you you took a large loss in your portfolio this is not a normal distribution it's just to realize loss it takes a long time to recover yeah so that's a huge convexity only go to one lunch in one run atomic right you got it you take that huge loss it takes a long time to recover or if you sit around you know and you miss the big gain it takes a long time to

recover you know so that it's the fundamental difference is that we have to think about those convexity costs and what they're doing to the compound return what do you think of products that are that are out there that claim to the so-called target date funds that that claim to take into account that the years I have until a retirement for example and therefore reallocating my equity bond split I'm a very old man at this time and they tell me that I should be involved now maybe in 1980 if I was a young man at that

time and I did invest in bonds you know there's an asymmetry of the fact that the bond returns could be very big as well and nowadays it tends to me that there could be a lot of pay losses in bonds so the risk of bonds today might be far different from the risk of bonds and 90 getting given a current low interest rate environment correct so the ideas why I I think that this is a stupid thing because what it's saying is that well you really want to do is say as you get older maybe

your risk appetite Falls because your human capital Falls but the ideal portfolio should take out of risk not bonds versus stock and the target date funds would say when you're young you should invest in stock when you're old you should invest in bonds it's not the correct model the correct model is risk when you're young what risk do you want to take and what is the risk as a function of your realized return you know what is the risk you want to take as to what has to do with the other parts of your of

your human capital other parts of your wealth structure and so the target date funds which are sort of stylized ways in which a thinking of numerical asset allocation are not taking account of what we should be accounting for what's the risk and how is the risk changing and what is the dynamics of risk you know nothing to do with forecasting weather returns are gonna be great or not just efficient risk management and a new target date front of the future will be a risk managed fund not a target date fund not a fun and not

one is saying I have a ten year rise in her three or as I've heard so many people say I have a 20 year horizon therefore I should run my portfolio differently from one year horizon or a six-month horizon that's not true we need a new way of looking at we have a new we have all the ways we have all the ways today are available we're just not focusing correctly we're not focusing on what we should be why because this relative value performance has taken over everyone says the bet you know the benchmark is

king if the benchmark is king and we just looked at relative performance when I developed her involved that was Mike Jensen and others who develop performance measurement it was just to say how can we say this manager is uh performing the benchmark that doesn't mean he says-- uh performing the benchmark or not outperforming the benchmark you should forget about the benchmark it's stupid what advice would you have for for typical investors I would like to see those who have skills or managers you know start dividing defining the portfolio that I'm describing and offer that to

investors is a way to think about this is then the investors can choose different levels of risk different levels of drawdown they can have and different stock demux then have that as the way to run a portfolio not ignoring this entirely something that's more dynamic has to be dynamic changing that's the be dynamic thank you well myron on behalf of investors everywhere i want to thank you for taking the time to share your thoughts with us you