Episode 8: Hari Krishnan [Doherty Advisors]

Profiting after market shock, Common Hedging Mistakes and Machine Learning in Markets.

In this episode, we chat with Hari Krishnan of Doherty Advisors. Hari is a volatility macro-focused hedge fund manager and the author of The Second Leg Down, a look at practical approaches to profiting after a market shock.

We start the conversation with Hari by talking about the academic theory that suggests hedging is not a good long-term strategy and how his experience of the practice differs from what theory would predict.

We then get into Hari’s view on ETFs and the potential systemic risk they pose to markets more broadly, and, what investors can do to manage that risk.

We look at common mistakes investors make when hedging including hedging too literally and too reasonably. Hari believes that investors typically buy options for scenarios that seem plausible when they should focus on how options will be repriced as volatility unfolds.

We then go into hedging strategies for different parts of a crisis, what Hari calls the first and second leg down. Hari has found that strategies that looked cheap before the first leg down tend to look expensive after and investors should adjust their hedging strategy as a result.

Finally we go into the role of Algos and machine learning in markets and how Hari’s views there have changed over time. You can find out more about Hari by buying his book ‘The Second Leg Down: Strategies for Profiting after a Market Sell-Off‘.

Listening options:

• Listen on Stitcher
• Listen on iTunes
• Listen on Spotify
• Android users can grab our RSS feed here.
• Listen on YouTube 

Have comments about the show, or ideas for things you’d like Taylor and Jason to discuss in future episodes? We’d love to hear from you at info@mutinyfund.com.

Transcript for Episode 8:

Taylor Pearson:
Hello and welcome, I’m Taylor Pearson and this is the Mutiny Podcast brought to you by Mutiny Fund, a multi-strategy long volatility fund designed to give retail investors a way to ensure their portfolios against volatility, tail risk and black swan events.

Taylor Pearson:
This podcast is an open-ended exploration of topics relating to growing and preserving your wealth, including investing markets, decision making under capacity, risk, volatility and complexity.

Taylor Pearson:
This podcast is provided for informational purposes only and should not be relied upon as legal business investment or tax advice. All opinions expressed by podcast participants are solely their own opinions and do not necessarily reflect the opinions of RCM alternatives, Mutiny Fund, their affiliates or companies featured.

Taylor Pearson:
Due to industry regulations, participants on this podcast are instructed to not make specific trade recommendations nor reference past or potential profits and listeners are reminded that managed futures, commodity trading, Forex trading and other alternative investments are complex and carry a risk of substantial losses.

Taylor Pearson:
As such, they are not suitable for all investors and you should not rely on any of the information as a substitute for the exercise of your own skill and judgment in making such a decision on the appropriateness of such investments. Visit www.rcmam.com/disclaimer for more information.

Taylor Pearson:
In this episode, we chat with Hari Krishnan of Doherty Advisors. Hari is a volatility focused macro hedge fund manager and the author of The Second Leg Down, a book that looks at practical approaches to profiting after a market shock.

Taylor Pearson:
We start the conversation with Hari by talking about the academic theory that suggest hedging is not a good long-term strategy and how his experience with the practice differs from what theory would predict.

Taylor Pearson:
We then get into Hari’s view on ETFs and the potential systemic risk they pose to markets more broadly and what investors could do to manage that risk. We look at common mistakes investors make when hedging, including hedging too literally and what Hari calls hedging too reasonably and we then go into hedging tradings for different parts of a crisis.

Taylor Pearson:
What Hari has found or what he talks about in his book The Second Leg Down is the strategies which look cheap before the first leg down tend to look expensive after investors adjust their hedging strategy as the crisis unfolds.

Taylor Pearson:
Finally, we go into the role of algos and machine learning and markets and how that’s evolved over the last 10 years and Hari’s view on how investors think differently by participating in markets increasingly influenced by machine learning. I hope you enjoy the show.

Taylor Pearson:
Hari, you’ve talked about sort of analogy between sort of financial crises and earthquakes. How do you think about that analogy and where do you see sort of that the limits or where it breaks down?

Hari Krishnan:
Well, every crisis is a little bit different. That’s point number one, but there are some similarities and the similarities tend to be these. Crises tend to occur when there’s too much credit in the system, too much leverage, people are borrowing too much and when people are overextended in certain markets.

Hari Krishnan:
So in 2008, it was in mortgages. It’s been various and people have been overly optimistic in various places over the years and that combined with a contraction in credit has proven to be toxic and there have been numerous crises over the years, there was the savings and loan crisis about 30 years ago.

Hari Krishnan:
Then there was stuff with the dot-com crisis which was again over leveraged, the credit crisis in 2002 and so on and so my view is that it’s a combination of positioning and too much leverage that’s now being cast that causes almost every crisis and I tend to call those things market tremors.

Hari Krishnan:
I don’t know if that’s the best phrase, but the idea is that even if there isn’t a lot of realized volatility in the market, there can be a lot of risk under the surface and that is something I look at to.

Taylor Pearson:
Just one of the things I’m curious about, you’ve talked about sort of how academic theory looks at hedging the line I pulled from your book is that hedging must have a negative risk premium as it reduces the non-diversifiable risk.

Hari Krishnan:
Oh yeah.

Taylor Pearson:
In your portfolio, what … I guess what, if you could unpack that statement a little bit, what does that look like in in practice and then maybe are some of the assumptions there and how would you sort of pick that apart or what sort of parts that you feel like do hold and maybe which parts of it don’t?

Hari Krishnan:
Okay. Well, I’m not the expert on this, but I know the basics and I’ll just repeat those. The basic idea is that if you invest in risky assets, at least according to theory, you should be compensated for that with a return and that applies to investing in government bonds versus investing in companies and it also applies in terms of where you invest on the capital structure.

Hari Krishnan:
So if you invest in bonds, you have more protections in place. So in theory, you should earn a lower return over the long-term. So academic theory has been bogged down if that’s the right phrase in the idea of risk premia for many, many years and if I’m not mistaken about 20 or 30 years ago, there was an emphasis on trying to find risk premia other than just being long the markets, being long in the US equity market.

Hari Krishnan:
The first realization of that was probably the so called Fama–French factors and those included things like being long small caps, overweight, small caps, being overweight companies that had a low price to book value so they’ve been trashed a little bit in price terms, but had some intrinsic value and more recently, it turned into things like being long momentum.

Hari Krishnan:
In other words, momentum is a strategy that tends to have a risk premium at least over time and it’s extended to other things like selling volatility, investing in distressed assets and you can go down the line.

Hari Krishnan:
I won’t belabor that point too much, but if you believe that theory, if you believe that taking on added on diversifiable risk which is kind of one of the cornerstones of cap and other theories has a positive risk premium, then anything that lays off that risk, that cuts the downside, in theory, and I’m using heavy quotes here should have a negative risk premium because you’re basically cutting the risk, the systematic risk in your portfolio in a way that would presumably cut into the risk premium that you were collecting everywhere else.

Hari Krishnan:
So that’s the idea. So in theory, if you believe that buying equities has a positive long-term expected return which I do in general, then anything that you do that’s mechanical to hedge that risk should have a negative return and it’s well known that insurance companies if they are profitable businesses and to be profitable because insurance is statistically overpriced.

Hari Krishnan:
And if that’s true that options on average over the cycle should also be statistically overpriced, but my arguments are typically that there are times and places where you can find value and that’s what I focus on.

Taylor Pearson:
Another way of looking at that too would you agree that like in isolation, those theories are correct, but the combination of those two factors might, it improves your compounding return over the business cycles?

Hari Krishnan:
Absolutely. Absolutely. You can take strategies that have a negative expected return, not that anybody really likes looking at the line item of those, but that dramatically improve not only the Sharpe ratio or the risk adjusted performance of another strategy, but also improve the overall return because if they cut the downside enough, you get better compounding benefits sometimes even with a strategy that doesn’t have the positive expected return.

Hari Krishnan:
You have to blend it. You need other stuff in there that does, of course, but if you can cut the big drawdowns at a reasonable cost, then you’re in business because you can ride out a lot of those sell-offs that other people would be forced out of the market from.

Taylor Pearson:
How do you think about sort of identifying those times when it does make sense to buy insurance? Now, one of the analogies we’ve talked about is you can’t predict the forest fire per se, but you can see there’s different factors, you can look at the wind speed and how much water is … How much rain has fallen in the last month and you can kind of … It’s more or less likely they’re wasting it like that or how do you sort of think about when [crosstalk 00:09:05] to buy insurance?

Hari Krishnan:
Okay. Yeah, there are two things. There is valuation and there’s a catalyst. So you could be a deep value buyer of insurance and one of the nice things about the in the options markets is that investors do at times get overly complacence and typically at the end of the credit cycle, what happens is that prices are going up for things like the S&P and a lot of people are worried that they missed the boat.

Hari Krishnan:
So they want to get involved, but to make themselves feel a bit better, perhaps they wait for little dips and they buy the dips and then any little dip is met with buying and when that happens, basically what what occurs is the volatility or at least the downside volatility of the S&P goes down.

Hari Krishnan:
So it looks less risky even though more people are piling in, there are more leveraged investors and there are more weak hands that have just gone in at the end who will be forced out if conditions worsen.

Hari Krishnan:
So that’s one way to think about when options are going to be cheap, it’s typically when people are overly exposed to those markets which is a bit counterintuitive, but that’s a good time to invest over the cycle, late in the cycle.

Hari Krishnan:
The other thing is looking for a catalyst, the most primitive catalyst is to look for some price action in your favor. So if the Australian dollar is trading at a low insurance premium, at a low options premium relative to the US let’s say, but it’s starting to turn in price terms, you still might be able to get in, buy insurance and take take advantage of the fact that there is a bit of a price catalyst that’s working in your favor.

Hari Krishnan:
That’s the most basic way to do it. If you want to get more sophisticated, you have to look at the structural risk embedded in the system and I can go into that in detail too, but the idea is that if too many people are overly aggressively positioned in a market and or there are passive or structured products that are trying to mine the risk premium in a market to the nth degree, there’s probably a lot of risk in there that isn’t being felt in price movements. That’s the idea.

Taylor Pearson:
Yeah, let’s go into it in terms of like structural risk beyond … How do you think about that? Or how do you feel like investors should think about structural risk?

Hari Krishnan:
That’s a tough one because it’s a little bit episodic, meaning that each time there’s a crash, it tends to originate in a different market. So it’s hard to screen for everything, but nowadays, the way I would think about it is if you can find a market where ETFs are dominant or especially mechanical trading ETFs are dominant, or ETNs, or a market where there’s a huge amount of structure products activity, then it’s probably ripe for some potential risk down the road and what we’ve got now is a situation that most investors are aware of, but really should worry about, which is that there are many dangers with low yields globally and one of the main dangers is that people get hungrier and hungrier for assets that have a yield.

Hari Krishnan:
It’s saying nowadays which everyone or many people know is that perversely, people buy equities for income and bonds for capital appreciation nowadays which is turning the whole investment equation on its head, the whole thing on its head and that’s just part of the search for yield.

Hari Krishnan:
So I would argue that persistently low interest rates are also a trigger for certain types of excessively yield hungry activities so you can look at rates, you can look at credit supply in the markets, whether its central bank, balance sheets, the amount of debt in the corporate sector and so on or you can look at price action as a catalyst, so the various once. Yeah.

Taylor Pearson:
You mentioned sort ETFs and ETNs and I think the popular perception of at least ETFs is like, “Oh, these are great, it’s part of the passive low fee trend, these are very low fee insurance.” What do you what do you see is sort of the risk there or how do you think about that risk?

Hari Krishnan:
Some of these ETFs are okay. Probably the spiders are fine and some of the big cap.

Taylor Pearson:
Yeah.

Hari Krishnan:
US and global ETFs are okay and the reason they’re okay is because there’s no liquidity mismatch between the shares that underlie the indices, and the ETFs themselves. They kind of trade in tandem.

Hari Krishnan:
If anything, the underlying shares could be more liquid in certain cases, that’s good because that allows the arbitrage process to play its role. In other words, if the ETF goes out of whack with the index, then the dealers in the ETF can step in and squash that spread.

Hari Krishnan:
They can put things back in line. That’s good. In other markets, it’s a mess and the reason it’s a mess is because partly because the incentives for ETF providers are very different even than they are for the vanguards of the world, the passive, the mutual funds that focused on passive investments.

Hari Krishnan:
ETFs ultimately don’t really care about performance, you take out the intelligent ETFs. I know that they’re implicitly dependent on performance, but I don’t think that in general, they feel the same reputation risk from having products that don’t do well as active managers do and so their strategy is just to throw stuff up against the wall and see what sticks and they will run with any number of ideas.

Hari Krishnan:
Now, there is a barrier to entry which means that the ETFs needs to get enough assets in the first year or so to pay its costs, otherwise, it won’t survive. But I think many ETF providers, especially outside of the big names, they’ll throw up anything and so a lot of these strategies are not well thought out and they’re not necessarily even designed to be well thought out.

Hari Krishnan:
We’ll see how it works and move on, but the danger is that in certain markets, you have problems whether it’s a problem of leverage which was the case with the volatility ETNs, the notes or with certain types of corporate bond ETFs where the bonds don’t trade and the ETFs apparently have fairly liquid intraday trading until they don’t where the arbitrage mechanism is uncertain at best and I perceive a lot of danger in old, but the biggest most liquid ETFs where the underlying is also very liquid.

Hari Krishnan:
Other than that, I think it’s a very dangerous space and you can even make less toxic arguments than that. For instance, basically what’s happening is that ETFs are coinciding with the rise in passive investing and passive investing has various dangers associated with it. Some of them are philosophical.

Hari Krishnan:
It’s kind of a threat to capitalism in a way if you’re a free market guy because money is no longer necessarily being steered to productive enterprise, if everybody buys the index, whatever was big going into the passive revolution will stay big and so there won’t be anybody underneath to say, “Oh, this company’s doing something important. It might be a bit smaller. I’d like to divert some money into that.”

Hari Krishnan:
If everybody went past if you had a horrible, horrible situation, I’m writing stuff about that now. So I’m scratching my head.

Taylor Pearson:
Just to put a finer point on when you’re talking about the illiquidity to liquidity delta of let’s say corporate bond ETFs. If somebody’s listening and they’re like, “Well, I don’t care. I don’t invest in corporate bond ETFs.”

Taylor Pearson:
What you’re looking at, to use your term, market tremors is your idea as if you had a violent unwind in these illiquid ETF bond portfolios, that can send tremors or correlations going to one across markets? Is that the way you kind of look at it?

Hari Krishnan:
Oh definitely because if corporate bond spreads blew out, this would have an issue across the board because throughout markets because what’s happening now, and again, others have commented upon this in more detail than I have is that repo markets are dominating the financial landscape and it’s not important to know exactly what that means other than to say that traditional bank lending has been replaced by collateralized lending where if you want to borrow money, you need to post some assets into that borrow. So people have a claim over those assets if you default.

Hari Krishnan:
Now, corporate bonds are increasingly used as collateral for borrowing money. Now, if corporate bond spreads blew out, all of that collateral would come under question and I think a lot of the lending markets would suffer as a result.

Hari Krishnan:
In addition, what’s happened recently is that for some years is that corporations have been issuing a lot of debt at low rates, and then partly using it to buy their shares back which has created a contraction in the supply of shares and since many institutions need to have some stock exposure that bid up the prices of shares partly as a function of that, all of those mechanisms would go or would be very vulnerable if credit spreads blew out.

Hari Krishnan:
So a lot of the lending market behavior or securitized lending that companies are doing could get stuffed if credit spreads go out.

Taylor Pearson:
I’m just going clarify on the just to make sure I understand sort of the fundamental thing with ETFs then is when the whatever is underlying the ETF, like we said the corporate bonds are less liquid than the ETF itself.

Hari Krishnan:
Yes.

Taylor Pearson:
You can have a scenario where there’s a dislocation in the … Or what sort of … How does that play out? Or how does that … What’s the sort of mechanism there?

Hari Krishnan:
Okay, there are three major players I guess, well a few major players in the ETF markets. One of them is the issuer person who builds the ETF gets the prospectus written and holds the assets underlying the index.

Hari Krishnan:
Then there are dealers, typically people who have institutions that have experienced trading the assets in the index. So for bonds, it might be some of the big investment banks and some of the big bond dealer houses, broker houses.

Hari Krishnan:
And basically, the ECF trades in the market, it has an updated price every 15 seconds or so and let’s say that the ETF just dropped, it had a flash crash for whatever reason, maybe something correlated to it went down quickly and then the algos stepped in and started selling the ETF.

Hari Krishnan:
In theory, the dealers would be incentivized to buy the ETF shares at a discount, hand them over to the provider and then receive a basket of representative bonds in return. Now, if the ETFs shares were trading at a discount, then there would be an arbitrage in theory because they would have bought the shares at a lower price than the value of the bonds that they receive that they can then sell into the market, but my argument is that these bonds would not sell.

Hari Krishnan:
There just isn’t sufficient liquidity in these markets anymore given that the bank dealer community has changed its pattern of behavior to support that sort of move. So the ETF could freeze or still there could be a feedback loop where they try and sell the bonds into the market, the bonds price down significantly and then the ETF has to adjust down again too.

Hari Krishnan:
So the arbitrage mechanism is basically supposed to make sure that the ETF trades in line with the index or with the reference basket, but the problem is that it requires that any mispricing is met with dealers stepping in either taking delivery on bonds or selling bonds back to the provider, and then doing a spread trade. So there’s a hidden tail risk in that bid.

Taylor Pearson:
You mentioned, Jason bought up the idea that being able to … That becoming almost a systemic risk that could spread from corporate bond ETFs to other parts of the market. I’m curious how you think about the existence of a systemic risk.

Taylor Pearson:
I guess I think about like myself in 2008, at that time, very little, if any sort of financial understanding, it was like very perplexing to me that this thing happened in residential real estate in America and everything everywhere went down as a result.

Taylor Pearson:
It wasn’t intuitively obvious, like that’s how markets work. How do you sort of think about like that? Is it similar to dynamic or how do you think about that dynamic and just sort of maybe how systemic risk is changing over time or how it has changed over time?

Hari Krishnan:
Probably the systemic risk of a small pocket of the markets driving a total meltdown is lower. Never say never, but I would argue that that’s the case because part of the problems with these ETFs are actually good things that are happening at banks, namely that banks are no longer warehousing large positions, they’re more careful about counterparty risks.

Hari Krishnan:
A lot of the credit default swap market has been transferred to exchanges at least for clearing. A lot of the regulation has been favorable in terms of reducing the opaqueness or the confusion as to who owes who what.

Hari Krishnan:
I mean, Lehman, I don’t even know if Lehman has been completely unwound now. It’s been over 10 years and that’s simply because the counterparty risks were so hard to unravel.

Hari Krishnan:
The derivative exposures had so many different players involved that it took ages to even understand what was wrong. Now I don’t think the problem is as big in that sense, but the problem of over leverage and overstimulation is worse.

Hari Krishnan:
So now we have the problem that is much more traditional in terms of its basis in credits over extension of credit and but has different wrinkles given the coordination across central banks, and all of that stuff.

Hari Krishnan:
So I don’t think we have as much systemic risk in the banking system as we did, but we have more overextended entities at large than we did before. So I think it would be less … It would be more understandable or easier to trace what would happen in this crisis than the previous one, but it could be quite bad as well.

Taylor Pearson:
You mentioned sort of like central banks and the wrinkles in there. How do you sort of look at that dynamic?

Hari Krishnan:
Well, the common perception is that central banks are just easing like crazy and it’s a race to zero which has already occurred in the yield curve and that all the stimulation isn’t working in terms of getting inflation to pick up and so on, but it does create certain types of inflation.

Hari Krishnan:
One has to define what is meant by inflation. Is it currency inflation? Is it high street … Main Street inflation? Is it the inflation in asset prices? What we’ve actually seen has just been a big inflation in asset prices.

Hari Krishnan:
It hasn’t necessarily been good for the average person, but it’s definitely encouraged asset … Major asset class speculation and duelists and I don’t know, I think in 2006, if around 2006, Alan Greenspan said that he had solved or he and his team had solved the problem of the economic cycle. Some years, a couple years before ISIS.

Taylor Pearson:
That was nice of them. That was helpful.

Hari Krishnan:
Yeah, they solved it. Yeah. They had solved it and now the central bankers are more humble in what they say, but certainly not much more humble in the degree of aggressiveness with which they’ll act from time to time.

Hari Krishnan:
They presumably believe at least in … The ECB believes that they can do firefighting on a consistent basis, they can just take any major risk off event and flood the markets with credit and solve the problem. So if anything in terms of actions, the hubris has gone up even though the stated confidence has gone down.

Taylor Pearson:
Even though in essence, we’re kind of also talking about negative yielding bonds and this might be somewhat tangential, but I think you touched on it, and an issue I’ve been thinking about a lot recently is that the idea of now we live in this world of collateralized borrowing or collateralized leverage, and so pieces and parts of that people may not realize is that negative yielding bonds can go onto your book as an asset that then you can use to collateralize borrowing and people don’t realize it’s almost a forced mechanism to buy those negative yielding bonds. That way you can lever up against them, is that too small of a market that I’m thinking about?

Hari Krishnan:
That’s a very good argument, that’s related to the repo [inaudible 00:27:19] that you can use these bonds, whatever their yield is as safe collateral and post that collateral and borrow against it.

Hari Krishnan:
The other thing is that some yield curves aren’t that flat even now. So you can make money just from the roll down. Again, that’s perhaps a bit more complex idea, but you can make money off the price appreciation as time goes by assuming that the yield curve stayed the same.

Hari Krishnan:
So there’s a bit of that return in it as well, but I think your point is the main point which is that bonds are in huge demand now because they drive the ability to borrow. There’s no more borrowing just by signing contracts and having a goodwill agreement with a bank.

Hari Krishnan:
It’s now post something to us, that thing had better keep its price, but there’s enough aggregate interest in that stuff that it does hold its price or at least it has and then you just borrow off that. So yes, yes.

Taylor Pearson:
Coming back I guess more towards hedging and how you think about that and how to actually get more mechanically, I guess my reading of some of your what you’ve written is sort of … The two common mistakes you see investors make in terms of how to execute a hedge or either A, doing it too literally.

Taylor Pearson:
I own some Apple stock so I’m going to go buy puts on Apple and that can be I guess in your view, sort of overly expensive or they do it almost too reasonably that investors buy options for moves that seem awful now, but in the midst of a severe bear market actually don’t look that bad.

Taylor Pearson:
In both cases, you should have lead to overpaying so I was hoping you talk through how you arrived at those two conclusions and how you think about it differently.

Hari Krishnan:
Okay, well, in the Apple case, if you were long Apple and you could trade out of it, but you decided to hedge it, you’re undoing your basic best in a way because if you like that specific name, hedging that specific name is going to cut into return quite dramatically and it’s a direct hit that you’re going to take.

Hari Krishnan:
So that in a way can be unproductive. I mean, my focus although it’s different from others and there are other approaches is to hedge against the systemic risk event to say, “Well, if things get bad enough, everything’s going to go down. So let me just hedge it in a relatively efficient way against a widespread crisis.”

Hari Krishnan:
And there’s a lot of evidence that when fear goes up in one asset class, it goes up in others and I’ve given some evidence for that in the first book. The second thing was this reasonableness idea.

Hari Krishnan:
Well, if you’re an options trader which I have been for many years, you don’t necessarily rely upon holding an option until expiration to make money. So if you buy a put on something on the S&P or the footsie or whatever and the S&P is trading at 3,000, and you buy a 2,500 puts with a few months to go, you don’t need the market to trade down to 2,500 to make money.

Hari Krishnan:
All you need is a repricing of risk at that strike. If the fear kicks in, then that option will gain in value and you can sell it and maybe buy a different option or do something else. So a lot of people would sit around the table and say, “Well, what do you think is a reasonable bad scenario for the S&P in the next six months?”

Hari Krishnan:
One person might say 10% down, another person might say 15, one might say five and so on and then the dumbest thing you could do would be to average those bets. Now it’s fine to average expected returns, but to average downside, bad downside expectations tends to mute the truth … It tends to understate the true severity of a real trickle-down. So that’s the problem and so what happens is that institutions often hedge too close to at the money which means if the S&P is at 3,000, they might hedge against a 10% drop, so they might buy 2,700 puts 10% down whereas if someone else were to buy 20% out of the money puts, not that those do or do not provide good value, but let’s just say that someone did it, those puts would cost a lot less in terms of premium outlay per unit than the ones that that had higher strikes.

Hari Krishnan:
But if the market traded down 10% very quickly, you wouldn’t make that much less because volatility would reprice enough that you’d get a kicker on the 20% down insurance claims that you had bought.

Hari Krishnan:
So, well more like a 15% down, but still. So the idea is that you don’t need to buy strikes that you think will … The market will go through in order to make money, that’s point number one and point number two is that if the market does drop 10% and you have 20% out of the money insurance, people will suddenly put that 20% downside scenario into play.

Hari Krishnan:
They’ll be worried about it. So if they didn’t worry about it at 3,000 they didn’t worry about it 2,400 S&P 3,000, some people will be worried at 2,700 if it’s quick enough, that 2,400 is in play and in 2008, I remember people going on the TV and saying that a 400 S&P was very probable or likely or 350 S&P and this and that.

Hari Krishnan:
It may have gone down that maybe we dodged a bullet, whatever, but those crazy low values suddenly came into play when the market was melting down around 700, that would have been inconceivable even at a thousand or 1,250 or whatever.

Hari Krishnan:
So buying these way out of the money strikes, if they’re not overpriced, is a great strategy because a lot of people think, “Oh, I’m not protecting against a terminal event scenario at expiration.” But I’m saying you don’t need to do that, you need to protect it against a repricing of fear in the markets and that’s the way options traders think and I think it’s well applied to hedging as well, make some money off changes in sentiment.

Taylor Pearson:
The idea with the 2,400 versus the 2,700 of the S&P at 3,000 is maybe the, I don’t know, I’m making these numbers up, but the 2,700 is a $2 option and the $24 there’s a $1 option, but if you have that sharp move down in the first month of a six-month options contract, that $1 or 2,400 might be $2 or $3. And so that repricing is where you see.

Hari Krishnan:
Yeah, absolutely and in a quiet market, the differential between the 2,700 and 2,400 might be five times different or 10 times different if the option doesn’t have too far to go. So you can actually stockpile even more of those things and put them into play.

Jason:
This is my … Just to re-highlight. It’s my favorite thing that you … That we’ve talked about in the past is this repricing and it’s almost impossible to talk about cheaper, expensive in option pricing because it’s regime dependent for lack of a better term, and how much it really is a phase shift from risk on risk off, whatever systems or models use the price options during risk on are almost inapplicable to a risk off event because you have that phase shift and like you said, that fear can do crazy things with the supply demand and the bid offer on those options and like you’re saying, it’s a psychological repricing of risk that matters.

Hari Krishnan:
Absolutely, yes. Yes, and we have often compared ourselves to consultants in what we do and a consulting community deals mainly with institutional mandates, pension funds that needs a hedge, insurance companies and so on and maybe we’re begging ourselves up here, but we think that acts of managers who also offer solutions in principle should have an edge because we’re not thinking of just a terminal payout, but of using some of the ideas that we apply in the alpha generation piece of our business to hedging.

Hari Krishnan:
We might lose a bit, we lose a bit of edge by focusing on buying protection outright and so on, but we’re still able to be more efficient than the consultants. So I would argue that people who go into these sorts of strategies whether it’s us or someone else you might talk to is potentially better served in terms of efficiency than a large institution that uses a consultant to do their hedging.

Hari Krishnan:
So I think you may … Your audience may actually get better ideas from people like us or other people you might talk to than the big institutions would. So it is a relative edge from being a bit smaller.

Taylor Pearson:
And I think in lot of [inaudible 00:36:49], I think people talk about the VIX, [inaudible 00:36:53] reviews the VIX as a barometer of whether it’s a good time to hedge or not. How do you think about hedging with the VIX or maybe just talk about sort of the relationship between the VIX and S&P and sort of how you think about those two?

Hari Krishnan:
Well, the first thing I’d say is that hedging with the VIX is a lot better than hedging with bonds. Some people say, “Well, you can just buy government bonds and hedge.” Okay, you can do that.

Hari Krishnan:
It would have worked since the early ’80s I’m sure because in during most crises, bonds have gone bid. They’ve gone up in price. It may continue to happen in the current regime, but over the long term, there’s no really strong structural connection between government debts and the equity market.

Hari Krishnan:
Whereas for the VIX, there is. The VIX and the S&P are linked, they’re mathematically linked. The more the S&P wiggles and wobbles about, the higher the VIX will be. That’s just a structural thing because if that were not the case, then speculators could say, “Sell or buy options and then hedge them in such a way as to take advantage of the mispricing.”

Hari Krishnan:
So there has to be a structural relation between the VIX and the S&P. As I’ve said many times, the trouble is you cannot buy the quoted fix, if you could, problem solved. Just by the S&P, buy the VIX.

Hari Krishnan:
The VIX has a much higher volatility than the S&P, but if you did it at say 80/20, an 80/20 ratio or 90/10 ratio even, you’d cut off all the downside more or less and you’d have a much bigger return as well because of the compounding effects and that would be great and in addition to that, if you bought the VIX when it was low, even better because the VIX has never gone below nine.

Hari Krishnan:
Maybe it did intraday recently, but it’s never gone much below nine and it’s gone as high as 80 in 2008 intraday and theoretically in 1987, it would have been over 100. So your upside would be huge, but you cannot buy the spot fix and so what the ETN suppliers have provided is a very poor substitute which is basically buying and rolling the futures, that has a return of minus -99.999%, something like that since 2009 and the reason for that is just, it’s a bit more technical, but it’s related to having to roll the futures contracts from month to month and that is extremely expensive.

Hari Krishnan:
You’re constantly forced to buy high and sell low with about a 5% cost every month that you do it, and that just crushes the return. So, putting that into practice is difficult. I’m not here to give a specific sales pitch, but there are things that you can do to improve on rolling the futures and that’s something we do look into.

Taylor Pearson:
I guess one of the things I liked reading through your book is the idea of a first and second leg down in markets that in at different points in a crisis, there’s better and worse ways to hedge and that’s not sort of consistent throughout the time that you said that it’s better to buy all in hedge sentiment in the early phases.

Hari Krishnan:
Yeah.

Taylor Pearson:
First leg down and then second leg down to hedge directly as the crisis sort of gets worse. I was wondering if you could walk through to maybe an example of that, either historically or hypothetical and sort of how that would work or how you would think about that?

Hari Krishnan:
Okay. Well, the two extremes are easiest to deal with. In a very quiet market, insurance tends to be cheap. So if you buy it, you can buy it in a way that’s theoretically, it’s theoretically undervalued, at least relative to history.

Hari Krishnan:
So on average, if volatility mean reverts in a reasonable amount of time, if it goes back to normal levels, you will make money. In the teeth of a crisis, it’s more difficult because at least in equity markets and stock markets which your audience may be familiar with, the quality of returns from shorting in a bear market tends to degrade at bit relative to being long or buying in a bull market, but in other markets, that’s not the case.

Hari Krishnan:
So in bond markets and currency markets and so forth, so if you follow the trend in a bear market in a diversified way, you should be able to protect against further losses at reasonably low cost.

Hari Krishnan:
So when sentiment is giving you cheap insurance buy it, when sentiment is no longer giving you cheap insurance because markets are in a frenzy, don’t do that, do something else. Take the directional move head on, and don’t pay up for the insurance claim, just do it directly.

Hari Krishnan:
Now, you have to be careful as to how you do it, maybe you need to hire someone to do it for you. Maybe you need to focus on certain markets, but the theory the principle is when insurance is cheap, buy it, because that’s cyclical, when it isn’t cheap, do something else. Hedge some other way.

Jason:
And just to put a finer point on that because you helped honestly solidify our thinking on portfolio construction through these ideas in your Second Leg Down book and that in the first leg down, before like you’re saying the sentiment is cheap, so you buy those let’s say put options for example and the implied volatility is low given a risk on market.

Jason:
But as soon as you have that let’s say a quick 10 to 20% down move, implied volatility is going to expand, those options are going to become very expensive. So what you’re saying is you can then roll total you’d call delta one which really just means the futures market and that means you don’t have to pay up for that implied volatility, you can just have a one for one gamma on the way down if you just buy into those futures markets whether that’s on the S&P or proxy markets, and I don’t know if I said that quite as eloquently, but I’d like to, but or simply, but maybe somebody else could [crosstalk 00:43:12]

Hari Krishnan:
No, you said it well. Well, I’ll try and rephrase it in a rough way which is that if volatility is 10% that you pay for an option, then you start making money very roughly or let’s say one month volatility is 2%. Let me keep it simple.

Hari Krishnan:
You start making money if you buy at a 2% volatility if the market moves more than 2% of the month, that’s when you start getting more profitable. Whereas if you’re buying at 8% monthly vol, monthly standard deviation, you need four times as big a move to start going in the money with your insurance claims.

Hari Krishnan:
So you don’t want to be reliance upon very big moves to make money on your insurance. You’d prefer to go in when even a modest size move will throw you into the money. So that’s kind of the way I think about it.

Hari Krishnan:
In 2008 In October, the market hit … The VIX hit 80 and the market traded down significantly from there to March 2009, but volatility actually went down. So vol was so big that you would have needed incredible moves to justify buying options outright, buying insurance outright in the teeth of the crisis. So you had to do something else to make money in the bear market.

Jason:
And that’s why it’s imperative to use the futures because you don’t have to pay up for that implied volatility? And that’s why you want to look … You want to move from combat option positions to linear short futures positions.

Hari Krishnan:
Yeah, to put it in more simple terms even, if you sell the futures and nothing happens, you break even. If you buy an option and nothing happens, you lose, you lose the premium you paid. So you need big stuff to happen when you buy options in a very volatile market.

Taylor Pearson:
And I guess, I was thinking about it just in terms of the VIX, if the VIX is … If the historical average I think around 20 something like that, if the VIX is at nine or 10 and you assume volatility tends to mean revert, it should move up towards 20. So it looks kind of cheap.

Taylor Pearson:
But at 80, you expect volatility to mean revert over some reasonable time period after [crosstalk 00:45:30] volatility has to expand a lot to go from, to be profitable going from 80 to I don’t know, whatever, 100, whatever you have to do.

Hari Krishnan:
Yeah, 140% which is a number that I like is would require the market goes up or down 10% every day which is pretty … Our circuit breakers would kick in by then. They would have to just keep going up or down 10% every single day.

Hari Krishnan:
80%, it’s over 5% every day which is a massive move. Every day would be a white knuckle rollercoaster. Even a day that was a 1% up or down move would kill the vol well below 80% quite quickly.

Jason:
And then you alluded too real quickly about buying weekly options versus monthly or quarterly. Is that something you’re seeing structurally in that undulating vol surface and you have maybe pension plans or other structural providers that are maybe overselling those weekly’s and that’s where you like to buy those puts or do you use it still get monthly and quarterly’s?

Hari Krishnan:
I’m more focused on things longer than a week nowadays. Weekly’s are very good if you need to cover a gap in your portfolio and you don’t feel comfortable selling now, but you need to put a floor on your losses over the short-term.

Hari Krishnan:
So it’s a mechanical device that you can use as an individual investor to block out any unpalatable downside scenarios for your portfolio and give yourself time to reposition and then fight another day.

Hari Krishnan:
That’s the practical use buying weekly options. If you are a seller, you tend to be more of the mindset, “Oh, but the shorter the time to maturity, the quicker and option decays.” So I want to sell those, but that’s not without a huge amount of risk.

Hari Krishnan:
I would hardly recommend people to do it. I don’t mind if others do it because it makes these options cheaper, but it’s not a good long-term strategy in my view. If you want to sell options, there are better ways to do it than taking on massive jump risk.

Taylor Pearson:
One of the other things I’ve heard you speak about Hari is sort of the role of algos and machine learning in markets, how do you sort of think about how that’s changed over the last 10 years or so and how does that affect how you look at markets?

Hari Krishnan:
There’s many things a little bit more dangerous. The first debate that used to come around was the one about is high frequency an added benefit or a penalty to markets? And the corrects reply is both.

Hari Krishnan:
It does cut costs most of the time, but just like in any other market making business which is what high frequency has tended to be, at some point, the market makers drop out and so you lose liquidity, not because the high frequency traders are acting as negative liquidity providers, if that makes any sense, but because they’re dropping out of the market, and the market has been assuming that they would always be there.

Hari Krishnan:
So they give an illusion of more liquidity than actually exists. By the same token as things have progressed, what we found is that algo trading tends to result in more concentrated positions, in more confidence that people can get out quickly just by a sampling risk at higher frequencies.

Hari Krishnan:
They think, “Oh well, I can get out more quickly and if I have to, I can see market conditions changing quickly and so on.” I think it’s added to the crowding risk and the risk of a quick flash crash partly because of the HFT crowd and partly because there’s more commonality and strategies than there ever was.

Hari Krishnan:
I mean, if you’re looking for structure in markets and you don’t look at things with an incredibly fine tooth comb, one much finer than what I possess, you’re going to wind up doing stuff that isn’t that different from what some other sharp people are doing somewhere else.

Hari Krishnan:
They will have cottoned on to roughly the same thing. If you’re on a huge winner, a huge run, eventually, people will figure it out. Just inferentially they’ll see that, “Oh, this pattern of trading seems to be making money.” Or maybe the brokers will notice it and so on and so I think an emphasis on using systematic rules creates more crowding or concentration and so it does add to the risk of flash crashes, short-term technical blowouts that were not in place before.

Taylor Pearson:
I guess what happened in eviction in 2018, was that an example event or was that something different?

Hari Krishnan:
There are different explanations. The one I liked was a bit fancy which said that, “Well, at the peak of the exchange traded notes. The inverse VIX exchanged traded notes.” There was the XIV and the SVXY. Why they accounted if they had fully replicated the index they were tracking, they accounted for 20% of all the open interest in the VIX futures complex.

Hari Krishnan:
That’s bad because it’s well known from the research on hostile takeovers which sounds completely removed that if you buy 50% of the shares of a stock and you announced to the world that you’re going to do it, that results in a price impact of about 30%.

Hari Krishnan:
That’s quite a range of different numbers from the studies, but let’s say on average, it’s 30%. So if they had to do 20% of that, and you make certain assumptions about price impact, you’re looking at a follow through on a five point initial move from say 15 of another several points and so you could almost track how bad the follow through would be just from based on the fact that the exchange traded notes had to replicate the index.

Hari Krishnan:
Now, notes do have the freedom of replicating in whatever way they want and they can go and try and hedge against a spike in VIX on a short ETN using S&P options and things that even in those markets, the amount of gamma, this is getting a bit technical, but the amount of hedging that would have been required would have been so large, the market could not have supported it, it would have been bigger than the average daily trading volume in the futures to do hedging so the market could not support that either.

Jason:
Another thing I’ve heard you talk about with machine learning that I find interesting is like, was it I think Kasparov said it, it was a man plus a machine learning will be the future, not just machine learning and that like you were saying before, I’ve heard you talk about if say you take the last 10 years and your machine learning algo was just running off the last 10 years of data, then you’re going to be selling volatility.

Jason:
It looks like a winning trade all day long and every piece of data you’re going to get is self-reinforcing that you should be selling volatility. Another way to look at it maybe in that February 2018, VIX-mageddon, VIX apocalypse is that your machine learning algo would have shown there was excess demand on the call side of the VIX and so it said, “Look, we should be selling these calls.”

Jason:
But a human can have common sense as you would say, to overlook that and go, “There’s a reason why people are buying these calls on this VIX complex and it’s not a simple machine learning algo so it’s man plus machine because you need that experience in the markets for common sense to override why are people over buying these positions.”

Hari Krishnan:
Yeah. Yeah, I mean, there’s a little throwaway comment I often make which is that machines make fewer mistakes than humans, but humans tend not to make massive mistakes. In other words, human wouldn’t trade a million contracts instead of a thousand.

Hari Krishnan:
Machine probably doesn’t feel much difference between the two or any difference whereas machines are more accurate with just doing regular stuff repetitively so the human oversight isn’t needed.

Hari Krishnan:
You can put any number of filters in and perhaps it’s not so simplistic as trading thousands of times too many contracts, but there are other structural errors that can creep in if you don’t have the right constraints in your program.

Hari Krishnan:
So you do need human oversight. Even the fully systematic programs in 2008 for a period like that, even if they don’t change the relative size of their positions, they’ll turn the knob up or down of overall leverage, that can be discretionary as well.

Taylor Pearson:
Hari, I know you’re working on a new book on market trimmers because you give us a preview or where you’re sort of looking at and interested in there.

Hari Krishnan:
Sure. Well, the first book which was called The Second Leg Down, the goal was to say, “Well, typically nobody gives you a mandate to hedge, at least historically when markets are quiet.”

Hari Krishnan:
So assume that on my side of the equation which is as a hedger on my side of the table, you only get a mandate when volatility has going up. What sort of hedges can you put on that may not be optimal, but aren’t too bad?

Hari Krishnan:
So they protects, and they don’t overpay for insurance. That was book one. Book two was more designed to say, “Well, given the relative quiet in the markets since the massive doses of QE were introduced, I know there were various episodes in 2012 and so forth, but the relative quiet on average. What can you do to identify risky situations where the market isn’t giving you a signal that risk has picked up?”

Hari Krishnan:
If you can buy options or buy insurance or reposition your portfolio in those scenarios, you should have an edge because you’re identifying risk where not even the options markets are aware of it.

Hari Krishnan:
It’s different from saying say, before an earnings report, a lot of people will say, “Oh, stock volatility goes down before an earnings report.” But that’s true, but options volatility tends to pick up before well known events, maybe not binary events, but events that are known with a specific date to have a big magnitude on prices.

Hari Krishnan:
I’m going a bit further and saying, “Well, Even the options market hasn’t really caught on to the added risk. So can you go in and identify situations in specific markets where buying options is a really good idea because the market is collectively unaware of the added risk?” That’s what the second book is about.

Hari Krishnan:
So commercially, the idea is let me try and cover both scenarios. Markets are quiet, I don’t want to be scaring people every month which I think is a bit insincere for one thing because none of us know the timing of the next crisis, but by the same token, I want to alert investors to potential weak spots in the global market system.

Hari Krishnan:
That’s what the book is about really and so I’ll go into various things that involve network risks in the markets is one node, is one player in the network getting too big, that revolves around the ETF stuff and then I’m also looking at things where on valuing stock markets, not just relative to fundamentals, but relative to the amount of credit in the system.

Hari Krishnan:
This is not my idea, others have come up with this as well, but I’m putting it into practical terms in terms of hedging. So, a good description is a lot of people were scratching their heads including me as to why equities kept going up from 2009 up to 2017, let’s say, and even a bit in 2018.

Hari Krishnan:
You could make the argument that yes, while the prices of equities were going up beyond fundamentals, the supply of bonds was also going up at an aggressive rate and the supply of loans, making it more difficult for institutions that have strategic asset allocation mandates let’s say that have to hold 60% equities and 40% bonds to do anything other than top up their equity exposure.

Hari Krishnan:
So looking at equity valuations in the context of credit is one of the things I’m diving into and tried to ride it up. So yeah.

Taylor Pearson:
Well, yeah. We look forward to when it comes out.

Hari Krishnan:
I do too. It will be relief, so.

Taylor Pearson:
Well, Hari, this is great, thank you so much for your time [crosstalk 00:58:21]. Thanks for listening. If you’d like more information about Mutiny Fund, you can go to mutinyfund.com or better yet, drop us a message.

Taylor Pearson:
I am Taylor at mutinyfund.com and Jason is Jason@mutinyfund.com and we’ll get back to you, you can find us on Twitter @mutinyfund and I am @taylorpearsonme.