Economics & Growth | Monetary Policy & Inflation | US
Podcast Transcript: Mustafa Chowdhury on Mortgage Risks, Fed Shocks and Derivatives
- By Bilal Hafeez
- Sign-up to discuss on Slack
- 40 min read
This is an edited transcript of our podcast episode with Mustafa Chowdhury, published 3 March 2023. Mustafa is a rates guru and member of the research team at Macro Hive. Before this, Mustafa was the Head of Rates, FX, and Derivatives at Voya Investments, where he helped manage $40 billion of assets. Prior to that, he was a Managing Director and Head of US Rates and MBS Strategy at Deutsche Bank. And in the 1990s, he was Co-head of Asset-Liability Management at Freddie Mac, where he was responsible for managing one of the world’s largest fixed income derivatives portfolios and trading desks. In this podcast, we discuss how financial institutions manage mortgages on their books, how options are used to manage market risk, what triggered the sub-prime/global financial crisis, and much more. While we have tried to make the transcript as accurate as possible, if you do notice any errors, let me know by email.
Introduction
Welcome to Macro Hive conversations with Bilal Hafeez. Macro Hive helps educate investors to provide investment insights on markets from crypto to equities to bonds. For our latest views, visit macrohive.com and please make sure to subscribe to our podcast show on Apple, Spotify or wherever you listen to your podcasts. We also have a YouTube channel with great video content, so make sure to subscribe to that too. On top of that, we have a free weekly newsletter that contains market insights and unlocked content. You can sign up for that at macrohive.com/free. Finally, please share this podcast with your friends, I’d be super, super grateful.
Now, onto this episode’s guest Mustafa Chowdhury. He’s a true rates guru and comes with over 30 years of experience in the industry and we’re pleased to say he’s soon joining the Macro Hive team. Before Macro Hive, Mustafa was the head of rates, FX and derivatives at Voya Investments where he helped manage $40 billion of assets. Prior to that, he was the managing director and head of US rates and MBS strategy at Deutsche Bank. And in the 1990s, he was the co-head of Asset Liability Management at Freddie Mac where he was responsible for managing one of the world’s largest fixed income derivatives portfolios and trading desks. Now, on to my conversation with him. Greetings and welcome, Mustafa, I’ve been really looking forward to this podcast, so welcome to the show.
Mustafa Chowdhury (00:01:22):
Thanks, Bilal, happy to be here. I’ve been a big follower of your podcast for a while and I see this is a great opportunity for me to be part of this. Thanks again.
Bilal Hafeez (00:01:33):
And as you well know that we have this tradition where I like to ask the origin story of my guests. So can you tell me something about where you studied, what you studied at university? Was it inevitable you went into finance? And something about your career up until this point.
Mustafa Chowdhury (00:01:48):
Sure. My introduction to finance, as we know it, was in my graduate school, University of California in San Diego. I’ve always wanted to be an economist for a long time. I started out in graduate school with the aspiration to be an econometrician and so I went to the most mathematical and data oriented graduate school I could imagine at that time. And did my PhD in economics under… I was lucky to have two noble laureates as co-chairs of my dissertation. One was Robert Engle, the other was the late, Clive Granger.
Bilal Hafeez (00:02:37):
Oh, as in Granger causality?
Mustafa Chowdhury (00:02:39):
Yes. So Granger introduced the concept of causality in econometrics. Before that, it was mostly about associations rather than causality. So his contribution… I was like, let’s… and Engle introduced the concept of the ARCH models and between the two of them together they have many, many new ideas and concepts in econometrics. So I started working with both of them and while I was working on my PhD, mostly on the topics that were hot at that time, various types of time paring, parameter models in time series, there was a crash in the stock market in ’87. Well known as the Black Monday, at that time. And suddenly I got very interested in finance, in options because eventually it was clear that the whole crash happened because of what was known in those days as portfolio insurance, which is basically the replicating optionality with dynamic hedging.
And the crash was just to prove that replicating portfolios doesn’t always work, especially when tail events happen. So I got very interested in that. I got very interested in options modelling, options pricing, implied vols. And eventually did a PhD dissertation that was sort of statistical models of implied volatility and supervised by both Rob Engle and Clive Granger. So that brought me from pure econometrics to finance and my interests in finance, interest in options pricing and volatility, et cetera, remained for the rest of my career. So after I left graduate school, I took a job as assistant professor in the finance department at Louisiana State University, otherwise known as LSU in the world of basketball and football. So I taught for a few years options and futures courses and investment courses.
In those days there was a lot of innovation. That was probably the peak innovation decade in finance, especially options pricing and asset pricing theories. The Blackshaws had just come up with their models about 10 years before and there was a lot of improvisations going on in those modelings. And so I spent a few years teaching and then working on something that’s actual investing or making decisions, involving options to be more interesting. So I moved from teaching to Freddie Mac, which is one of the two GSEs that hold very large mortgage portfolios in the United States, in the world actually. Well, it’s all about optionality in the Freddie Mac and Fannie Mae’s business, which is basically they hold a very large portfolio of mortgages and print out the credit portion, which goes into the credit risk management of the portfolio. And then the interest rate portion of it, which is basically a gigantic leverage short position in options. So my job I spent about a decade-
Mortgage Management and Using Options to Manage Market Risk
Bilal Hafeez (00:06:18):
And just as an aside there, the reason that’s an option position is because people can prepay their mortgage. And so-
Mustafa Chowdhury (00:06:25):
Exactly-
Bilal Hafeez (00:06:26):
If you are a holder of a mortgage like Freddie Mac was, you’re never quite sure when somebody’s going to prepay their mortgage. And suddenly you had hedged, you would put on investments to manage the interest risk and suddenly that hedge… well, that position’s gone, but you still have a hedge, so you have to suddenly decide what to do with that.
Mustafa Chowdhury (00:06:43):
Exactly. So the homeowners who are taking on the mortgage can refinance and that changes the whole characteristics of the cash flows. So if you think about a typical mortgage in the United States, which is a 30-year maturity, if it’s never prepaid, you can have duration as long as 15, 16, 17 years. But a mortgage can also refinance in a matter of months since after it’s originated and you get a duration of like four months. So you can have the same financial contract that can swing between a duration of 15 years to a duration of six months just because of simple changes in interest rates. So that’s known as negative convexity of a mortgage portfolio. So whenever you have an option position, if you are short optionality, you are negatively convex, and if you are long, the optionality you are positively convex. And what Freddie Mac was and similarly with Fannie Mae, and those two were sort of unique in this, was a huge short position in a hugely negative convex bond position.
So my job was to hedge manage that convexity aspect of that mortgage portfolio. Basically over time as interest rates change, the duration of the portfolio changes and you do delta hedging to keep the duration in the same place. So the problem was this is a gigantic portfolio, the Freddie Mac’s portfolio, when I started was about $200 billion and when I left 10 years later was about $650 billion. And so given that it’s a gigantic portfolio, delta hedging a portfolio of that size also means massive trades in either buying bonds or selling bonds or receiving in swaps or paying in swaps.
This is all improvisation and I feel very lucky to be in that position to learn and improvise this whole hedging MBS. That was the height, peak decade of how much influence the MBS market had on the rest of the bond markets. So learning number one, that you cannot really delta hedge a portfolio of that size because it’s very negatively convex. As I said, the duration could be a 15-year or six months, in a matter of a few weeks it could change. But also the leverage of this GSE was monumental. These days, 10:1 leverage hedge fund we think is very risky.
The typical GSE was about 40:1 leverage on a book value basis and sometimes the mark-to-market leverage could be 100:1. So you got a massively leveraged portfolio that has negative convexity. So the combination of negative convexity and leverage created huge movements in the market just as the portfolio swings a little bit. It’s like if you move the Titanic a little bit, it creates a huge wake around the whole system. So one thing that I… and I feel very lucky to have that opportunity to improvise on a lot of that. That instead of delta hedging, introduce more optionality on the hedging side. And that’s one of the reasons that the swaptions were introduced. I was one of the first people that started using swaptions to hedge a leverage mortgage portfolio from my hedging desk at Freddie Mac.
Then there was this, my academic knowledge on that field and the difference between the practical application was huge. What is the strike price of a mortgage portfolio? What’s the maturity of a mortgage portfolio? You don’t know any of that, so you do trial and error. And eventually my thought was, maybe about a hundred basis points, the mortgages used to be, in the late ‘90s, about six and a half sevens in terms of average coupon in those days. It’s coming back there, but we’re sort of in cycles in a funny way. So firstly, as I started using swaptions, the big experiment was the strike. And as you trial and error and see, because there’s no mathematical model that will tell you what is-
Bilal Hafeez (00:11:27):
And just to be clear just for our audience, the swaption is simply an option on an interest rate-
Mustafa Chowdhury (00:11:32):
On a swap-
Bilal Hafeez (00:11:33):
Contract on a swap, so swaps the type of interest contract. And the strike is the price or the level at which the option would be in the money.
Mustafa Chowdhury (00:11:43):
Correct.
Bilal Hafeez (00:11:44):
Yeah.
Mustafa Chowdhury (00:11:44):
So a payer swaption would be a right to pay fixed. A strike would be the interest rate which you would have the right to pay fixed. So if you believe that the interest rates are going up, then you do a swaption where you have a right to pay fixed. If you have a belief that the interest rates are going down or you have exposure to lower interest rates, then you have a right to receive fixed.
So my conclusion was that if I include a lot of low strike, right to receive fix in the portfolio, because of the inefficiency of homeowners in refinancing, eventually by the time they’re refinancing large numbers, it won’t be until another a hundred basis points or more in interest rate decline. So it makes sense to buy lowest strike optionality.
So then I started introducing and they had buy-in from the top management of the institution that it makes more sense instead of delta hedging a gigantic portfolio like that. So added billions, initially the… it was a big learning process for the vol market, fixed income vol market at that time because SKU was more of a theoretical concept rather than actual practical.
Now that the GSEs started buying SKU, SKU it basically means buying an optionality that is not at the money or at the current interest rate but at a lower interest rate or a higher interest rate. So the demand for the SKU started going up in the financial system in the US. And before the market was even clear about it, because it’s a portfolio of five to $600 billion, the amount of optionality that had to be bought was also like hundreds of billions of dollars in notional.
So the nice thing about the ‘90s is because SKU was not understood very well, it was mostly priced, you pay a hundred basis point higher involved for hundred basis point out of the money of optionality, it wouldn’t fluctuate much. So that was one of my huge transitions from understanding academic optionality in my PhD work and my teaching days to actually load, I would use the word load up on SKU in the option market in the US. At the market-
Bilal Hafeez (00:14:14):
So you were really there during the beginnings of the swaptions market as it was developing-
Mustafa Chowdhury (00:14:19):
I started it in a big start-
Bilal Hafeez (00:14:20):
You started it, yeah. And then also at the same time you recognised and saw that there was SKU in the market, which in theory shouldn’t really exist, I suppose. At least it would be a symmetric level of vol on neither side of the strike or at the money strike.
Mustafa Chowdhury (00:14:38):
Yeah, the market was smart enough by that time, by the ‘90s, that there should be a lower strike, should be higher priced, but the market had no idea how much higher a price should this be. So as GSEs went and paid only a hundred basis points higher, it wasn’t high enough. And the whole street thought of this as an arbitrage opportunity by constantly selling SKU to the GSEs, especially Freddie Mac.
So the biggest test was eventually after the dot-com bubble burst, if you remember it, there was a sudden big collapse in the stock market and interest rates started falling very fast. So in a month there was some massive decline in interest rate in 2001 few months after the dot-com crash. And suddenly the six and a half mortgage portfolio became very in the money and all the swaptions, like $500 plus billion of notional became in the money when they were statistically supposed to be very out of the money given the volatility of interest rates in those days.
Volatility of interest rate was very low in those days, they realised volatility. And so sudden increase, the swaption book suddenly had a profit mark-to-market of $5 billion plus and it shocked the whole system in terms of how much losses the other side had to take because all this SKU being sold to the GSEs. For me, it was a huge proof of concept where, how do you hedge the negative convexity of a mortgage portfolio that is highly leveraged? But then the story goes on and I will continue with that. But there was a huge gain and then the management had to deal with the gain and there’s a whole different story, how the management dealt with the gain. But the learning process was that the negative convexity of the mortgages to manage it required a huge amount.
It’s just a theoretical understanding of optionality doesn’t give you that, it’s a huge trial and error. But since then interest rates pretty much either declined or stayed stable. And so the MBS universe has become more and more lower coupon and the convexity lower and lower. So the hedging with options have declined over time, and I also left GSEs after this massive outcome of the hedging, and moved to the sell side on the street. But that was the story of the beginning of swaption market, beginning of the huge total repricing of SKU in-
Bilal Hafeez (00:17:41):
And let’s cut back to that, to today’s world. But before we do that, just to finish up your career path, you then went to work for Deutsche running rates research there for a period of time. What was the big learning during that period? Because now you were managing risk in the previous… well, you went from academia theory to practise or managing swaptions risk, then you went to the sell side, which is kind of the other side of the fence, so to speak, in terms of financial institutions. What was the learning there?
The Subprime Mortgage/Global Financial Crisis
Mustafa Chowdhury (00:18:12):
Correct. So that was the story of my life, several years of theory and then years of practise, then go back to years of research and then go back to practise. That’s been the story for years. So then when I went to Deutsche, the timing was really interesting to be in the Deutsche Bank trading floor as the head of rates and agency and MBS research. Just leading up to the GFC, the subprime crisis and right after that. But from a research and client point of view, some key things were first, clients…
I’ll just give you some anecdotes. One of them was the whole ABX index and how that was created mostly by the non-agency mortgage desk and the non-agency research guys. But I spent a lot of time with the same team as the rates’ person working next to them. So see the whole thing evolve. Eventually, there was a movie, there was a book, all of that. But I saw that movie in real life in front of me making client presentations on part of the client presentations of ABX index-
Bilal Hafeez (00:19:34):
ABX is Asset Backed Index, which basically was an index that held a basket of asset backed securities on there. And that was probably a really important step into some transparency for the market because before that point you had all of these asset backed securities. Often the assets were mortgages, sort of subprime mortgages I suppose, and just assets within there. But it was okay because suddenly this index got created, which was I guess, was it ’06, ’07? Was that the time?
Mustafa Chowdhury (00:20:00):
Yes. Oh ’06, ’07.
Bilal Hafeez (00:20:03):
Yeah. And then suddenly people saw you were able to see a price for the aggregate market-
Mustafa Chowdhury (00:20:09):
Exactly. Before that it was not possible to short housing in the US. There was not a clear instrument, liquid instrument that you could use to short housing. So no matter how much you didn’t believe in the fact that home prices were skyrocketing in those years, in ’06, ’07 and going into ’08, you couldn’t really do a trade that would short houses. There was no market to short houses. So this allowed a way to short the housing market and that came from few of houses in the street, but Deutsche was key in developing that index and developing… so I spent a lot of time with the research team at Deutsche as the index was developed, making presentations to clients-
Bilal Hafeez (00:20:59):
And at the time Greg Lipman was at Deutsche Bank who was played by Ryan Gosling in The Big Short, the film.
Mustafa Chowdhury (00:21:05):
Correct, that’s exactly… and the film was very real. Watching the movie, I felt like it was as real as it got at that time because I saw that movie in real life in front of me. Eugene Shu, who was the research head of the non-agency research, spent a lot of time convincing clients to short the mortgage market with this instrument. And I worked very closely with Eugene and that also in presentations to clients as well. And so later on watching that in a movie and in the book was sort of surreal to me.
I saw clients that made a lot of money doing that trade and we have seen clients that did that trade and then unwind because the negative carry was too high and missing out on that whole opportunity. So I saw the whole success stories and failures. That other piece that was very interesting in Deutsche days where we used to have a very good team in New York that covered GSEs, which myself and we had Mike Mayo who used to cover the banking system who was also a very well known and really good bank strategist, he was at that time at Deutsche Bank.
And then Christian Memani who we used to do corporate bonds and three of us would go make presentations to clients about vulnerability of each one of the sectors, in the middle of the crisis. I have complete understanding… I would give totally transparent picture of how little capital the GSEs have for the amount of volatility that’s expected. And a lot of clients would say, ‘No, this is the biggest opportunity to buy,’ and so there was a lot of disbelief that these gigantic rock solid institutions are so vulnerable in the United States at that time.
So seeing those responses, seeing some believers, some complete disbelief that these rocks like GSEs, the banks that are massive and apparently solid could crumble like a house of cards. So that was another piece, that in my Deutsche days being in the research I had a better picture than if I were actually running risk at that time. So the timing was very good as well. So that was the Deutsche experience. So both sides of fixed income, both sides of extreme volatility, it has been a very rewarding career for me.
And then I worked, again back to risk taking in Voya Financial, which is a new name for ING, US investing in global bond portfolios as well as other fixed income. Also hedging the different kind of convexity which is the variable annuity insurance type hedging. Well, that’s also very negatively convex but different type of negative convexing. So that’s my career half of the time in the buy side, half of the time in risk taking, half of the time in doing research.
Bilal Hafeez (00:24:25):
Great. And now you’ll be joining Macro Hive as well. You’ll be-
Mustafa Chowdhury (00:24:29):
I’m really looking forward to it.
A Safer Fed-‘Nationalised’ Mortgage Market
Bilal Hafeez (00:24:30):
Yeah. Now, I did want to talk about markets today and the big question first is, given everything you’ve talked about the mortgages and housing crisis, is there a housing issue today at all in the mortgage market? Because obviously house prices have fallen a lot over the past year or so, but it doesn’t feel like we’re in the same phase as a subprime crisis. It doesn’t seem like the leverage is the same and so on. But so is there anything to worry about with housing and mortgages?
Mustafa Chowdhury (00:24:57):
I don’t think so. Leverage is… several things happened in the US mortgage market that made it fairly resistant to big crashes going forward and it’s not necessarily a good thing. One of them, is that the individual homeowners are a lot less leveraged or minuscule leveraged relative to free 2008. So balance sheets of individual households are very solid. Second, the banks don’t own as much of the mortgages as they used to own in those days. Most mortgages are owned by the Federal Reserve, which is a government entity and it used to be owned by GSE Freddie, Fannie and the banking system. So the government was part of the player. Now, the government is mostly a hundred percent owner of MBS.
And two things changed. The government is not going to collapse as we see in the Fed, no matter whatever happens to mortgages, there must have been a big decline in mark-to-market value. But the Fed, nothing happens to the Fed. Secondly, which is actually more interesting and could be potentially my research topic in the future, that the fact that duration of mortgages have increased massively in this Fed hiking cycle. All the duration is held by the Federal Reserve and this massive change in duration leaves the Federal Reserve unscathed. Is that a good thing, the nationalisation of the USA-
Bilal Hafeez (00:26:46):
It leaves them unscathed because they don’t suffer mark-to-market.
Mustafa Chowdhury (00:26:51):
They don’t care about mark-to-market. They don’t-
Bilal Hafeez (00:26:53):
Because obviously when you are very long duration you become very sensitive to jumps in interest rates. And obviously rates have gone up, so in theory, if they were a private investor with normal mark-to-market considerations, they’d be losing money or they’d have to do something to fix the problem. But in their case they don’t have to worry about that.
Mustafa Chowdhury (00:27:14):
Hugely. And that’s one, is that a free lunch that you can just rationalise the mortgage market then provide stability to housing market for decades to come? I’m not sure Fed will ever be able to unwind this mortgage book except little bits here and there that’s running off. Actually there is not much runoff going on but then if the US private sector is capable of taking down that kind of size. The only window of time that has happened was in the period between 2003 and 2008 where there was quite a bit of private sector mortgage creation and it didn’t end very well. The rest of the time, in the ‘90s, GSEs were the main players and they hedged the rates quite well, but they didn’t have significant… they still pretty much supplied the liquidity to American home ownership, so home prices didn’t really collapse. And then now that the GSE Federal Reserve supplying the liquidity to the mortgage market, it’s unlikely for home prices to collapse.
So what’s puzzling me now after the last one year of intrastate increases and increases in the mortgage yield from two and a half percent to six and a half percent. That’s a massive change if you look in the historical context. So what it means is that a large amount of duration changed hands from homeowners, house owners in the United States to the government in the matter of a year that’s been happening as the Fed’s raising interest rates. So the homeowners then have more capacity to take risk than before because their risk had declined. They just transferred it more to the Fed. And so homeowners can now go and borrow more. I think that’s probably what they’re doing, they have the capacity to borrow more for more consumption. They have-
Bilal Hafeez (00:29:25):
And just to be clear what you mean there is at the beginning when interest rates were very low, households had a lot of interest rate risk because there was the possibility of rates going up. But because they refinanced over the course of the last year, they’ve got rid of that interest rate risk because they’ve locked into 30-year mortgages.
Mustafa Chowdhury (00:29:42):
Precisely.
Bilal Hafeez (00:29:42):
They’ve given over the Fed, so the Fed is actually the one with the hot potato-
Mustafa Chowdhury (00:29:48):
Feds hold the duration risk. And so the Fed’s holding the duration risk, different chunk of the duration risk that has transferred from households to the Fed. So households now actually, and they may not be consciously computing their duration using some duration calculator. But subconsciously may still be driving a large number of people to behave more like what the model expects, although they’re not consciously doing it, that they see that their risk is low. In their household, general balance sheet risk is low because a large chunk has been transferred to the Fed. So given that they have low duration in their household balance sheet, they can now take on duration.
Bilal Hafeez (00:30:32):
Okay. And how would they take on duration? I mean, they could basically take on credit card debt or take a bank loan out or something or just take other forms of borrowing on.
Mustafa Chowdhury (00:30:46):
On the other side, they can do long-term investments than taking more risks with the assets-
Bilal Hafeez (00:30:54):
Oh, on the asset side-
Mustafa Chowdhury (00:30:57):
On the asset side. They can buy equities, they can buy the long-term bonds in their portfolio. And that may be keeping the yield curve more inverted than we have ever seen in a Fed scenario. In the past, the yield curve was kind of flat, for example, in 2004, ‘05, ‘06 as the Fed was hiking, yield curve was very flat and we kind of blamed mostly foreign Central Bank buyers. Especially China and Japan that were bidding on the treasuries, keeping the curve flatter than anyone would expect. Today, it’s not only as flat as that, it’s way more flat than that. And we know that there is some Central Bank demand, but it’s not that large.
We still have some Japanese demand but there is not a huge amount of Chinese demand going on. So there must be some other buyer, some other something else that’s causing the belly of the curve to be so much lower, and the curve so inverted relative to history. And I think that’s maybe because there is so much duration taking capacity in the system that risk taking, even in the face of historic Fed hikes, there’s still demand for risk out there. Whether it’s equities, whether it’s tech sos, whether it’s long end of the yield curve. So this is a part, and is part of my research topic, I’m thinking about this quite a bit because where is this? There must be a risk accounting going on in the system and the Fed’s taking out so much risk meant there must be too many others who don’t have enough risk.
How Fed Quantitative Tightening (QT) Impacts Markets
Bilal Hafeez (00:32:44):
Then how does quantitative tightening fit into this? Because then the Fed is reducing the size of the balance sheet. But is it then that because they’re holding auto mortgages and they’re just offloading the treasuries, it’s the mortgage side actually that has a bigger impact which then keeps overall yields lower? Because you would’ve thought if the Fed’s now rolling off its treasuries and they’re selling their treasuries or they will then surely long-term yields will start to go up and you get curve steepening, but that hasn’t really happened.
Mustafa Chowdhury (00:33:13):
Yes, the mortgage portion is very little that the Fed’s reducing the balance sheet, the mortgage portion of the… they keep saying it but they don’t have a way of reducing the mortgage portion without creating serious turmoil in the mortgage market, in the housing market. And they don’t want to take any of that risk and that leaves their treasuries. And even treasuries, the duration, if you look at the amount of duration that’s coming out, you have to look at it from the cash balance sheet of the Fed and then there is the duration balance sheet of the Fed. And then there is a convexity balance sheet as well and let’s not go there. But just keeping it simple, the cash balance sheet is going according to plan. It’s declining mostly by deduction in the reserves for the banking system because the treasury’s account is more or less stable, reverse triple, is actually growing.
So it’s mostly through the reserves in the banking system. That’s just the cash, but the duration is also not going down as fast as the cash is because it’s a small front and is more running off not the long end. So the Fed’s doing very little QT in a duration sense and Fed’s doing very minuscule QT in a convexity sense. It’s only doing cash sense and that’s what’s affecting mostly the reserves of the banking system. And each one has its own separate effect on the financial system, the latter two which is the duration in convexity is minuscule, it’s only the cash that’s declining. So, so far, the effect of QT is quite small and I think it’ll remain small.
View on US Interest Rates?
Bilal Hafeez (00:35:07):
Yeah. And I was going to then ask given we’re in this strange environment where as you say, rates have gone up, housing markets not under stress, so we can’t use the ’08, ’09 parallel. At the same time yield curve’s inversing more than we are used to, at least since the ‘90s and 2000s onwards. I mean, how would you position in these types of markets? We know Dominique for example, is calling for 8%, which is obviously a very aggressive view. Consensus is more like five and a half percent peak in the Fed. But given the uncertainty around the Fed cycle, I mean, as somebody who’s more kind of applied, how would you position in this type of market where you have these kinds of almost bimodal or trimodal outcomes?
Mustafa Chowdhury (00:35:53):
My view is that the Fed will be extremely risk averse in this direction of hiking because they lost so much reputation in hiking too slowly last in 2021 or starting to hike so slowly in the whole transitory versus permanent inflation. There has been a huge amount of dent on the reputation of the Federal Reserve. As a process, this direction around, the Fed will just be totally data dependent and as long as the data is strong, they will keep hiking because they won’t be trying to… ideally they’re supposed to be preemptive looking ahead a year from now because these are all lagging indicators, whether it’s inflation, whether it’s jobs, these are all lagging indicators. If you look at these numbers and making your policy, you are always late, but Fed will have no other choice but to do that because the pressure of making mistakes and in this direction is high. But that will actually lead them to make a mistake, in my opinion, and they will over hike more than probably is necessary. So I don’t know whether it’s 8%, 7%, 6%, but over hiking is quite possible. So that is-
Could the US Curve Invert More?
Bilal Hafeez (00:37:14):
So that suggests further inversion then, doesn’t it?
Mustafa Chowdhury (00:37:16):
That’s further inversion and that suggests that the long end would refuse to follow that. And we see that in the break-evens, the five-year break-evens, 10-year break-evens are fairly stable at somewhere around two and a quarter-ish, two and a half-ish percent. So the market believes that inflation in the long run will remain fairly anchored no matter what the Fed does, and no reason to rush to sell the long end of the curve. And then forget the breakeven, if you look at the long end real rates, I think that long end real rate could still be fairly anchored. And it does go up somewhat as soon as the Fed gives a signal that the dual high you can add 25, oh no, the peak rate is not 525, it’s really 550. And then a few months later we say, ‘Oh no, no, no, it’s 6% or six and a half percent.’
Wherever it goes the rear rate goes up by some amount without changing much of the breakeven. So I would just buy as these real rates go up. In fact, with this massive jobs data in February followed by the retail sales and the PCE, it’s an opportunity. If you see the real rate, five-year real rate went, oh, it goes up by a certain amount. And as they keep going higher you just buy some of that, take a position in the value of the real rate curve. That’s what I would do. It’s more like a buying opportunity and it’ll eventually probably invert more, but at the end the Fed will break something. If anyone’s thinking that this will be one exception in decades that Fed ends the hiking cycle without breaking something, no, it’s not going to be an exception. It’ll end ugly no matter how you look at it.
And when it ends ugly, that’s when the Fed’s start easing very fast. Every time there’s rapid easing and we’ll get back the curve correct. So all of this action, it’ll be in the front end, and will just be in the belly of the curve and accumulate for the, mostly I think, tips. Because if you believe that the Fed will maybe change its target from 2% to 3% or 4%, which I don’t think they will eventually. But if it does then it will help you with buying the eal buying tips in the belly of the curve because your carry from actual realised inflation would be very good. So my trade would be just to use these selloffs as an opportunity to accumulate bonds. And also remember that so much duration is held by the Fed, private sector duration is not much. So adding duration in the long end is not a bad thing, but front end is going to be crazy. There will be a lot of volatility and collapses, et cetera happening.
Why the Fed Will Break Something
Bilal Hafeez (00:40:26):
Yeah, okay. So most of our… I did notice you said that the Fed, when they hike, they always end up breaking something and this time will be no different. So what did you mean by that?
Mustafa Chowdhury (00:40:37):
I think the Fed’s always behind, I haven’t seen in my 30-year career in the last 15 to 20 years the Fed ahead of the data. The Fed’s always chasing. And so what happens because it chases and it goes longer in any direction than it should be. When it’s easing in right after the pandemic or previously it doesn’t know when to stop so it goes to zero and then it finds something new and it keeps doing it. And especially on the hiking direction, the same thing, we saw every hiking cycle they keep going until they’ve gone too far to create a recession or a systemic financial crisis. A smooth exit from hikes, I just don’t remember. And this time around it’s going to be even more because the Fed’s feeling defensive about making a major mistake in forecasting inflation in 2001, despite observing inflation, they keep saying it’s transitory for at least a year until rapidly changing their mind.
So right now they are not going to be proactive to stop hiking. For example, shelter inflation has a significant lag and the lag is a little longer lately than historically from actual new rentals versus shelter inflation. So as long as the data is strong, they will keep hiking and we know because of the lag structure that we have now, that data will remain strong for a while. So they won’t stop. They will just keep doing it and data will remain strong and they will keep hiking, in my opinion. And eventually, if not, I’m doubtful there will be a liquidity or financial crisis like some of the past end of Fed cycles. Like we saw in ’95 for example, ’94, ’95, we had all the big crashes in local governments like Orange County and several hedge funds.
Bilal Hafeez (00:42:59):
And then we also saw the EM Crisis as well. We had the Tequila Crisis in Mexico, then we saw Asia Crisis in ’97-
Mustafa Chowdhury (00:43:05):
Exactly-
Bilal Hafeez (00:43:06):
LTCM ’98 and so on.
Mustafa Chowdhury (00:43:09):
LTCM ’98. So they always crash. Either it’s a liquidity crash or an economy crash and most of the time there is some hidden leverage somewhere that no one knew about and the Fed basically pricks that leverage.
Bilal Hafeez (00:43:23):
But let me ask you, I mean, what’s the hidden leverage today? I mean, I know obviously it’s hard to know that before the event, but if you had to find sources of leverage in the system today, which could potentially be vulnerable, what would you say they are?
Mustafa Chowdhury (00:43:38):
Well, we know what’s not very levered, it’s the households, household seems to be in pretty good shape. Which makes me feel a little better, that crash may not be as big. But the obvious culprit is emerging markets, of course, we already are seeing that it started in terms of the funding issues, et cetera, going on in the emerging markets. But there has… I think that people are ignoring the margin compression in corporations. You don’t see that in corporate bonds, which are doing really well in all air spaces. High yield is doing really well, IG is doing very well. But if you observe the corporate market for post-GFC, the switch from equity to equity buybacks and replacing with debt has been phenomenal historically. And the margin compression that’s happening now with these robust wages, robust wages and also higher and higher interest rates, funding costs, it’s sort of in the back burner for everyone.
Size of the Fed’s Balance Sheet Losses
Bilal Hafeez (00:44:47):
So what you’re saying is profit margins, which are being at very high levels, they’re going to come under pressure and start to fall partly as interest rates go up. Interest rates costs go up, but also because wages go up as well. And then in the end that could be the thing that cracks the corporate bond market, perhaps.
Mustafa Chowdhury (00:45:03):
That seems like one possibility. The mortgage market is highly unlikely because we can talk about that as why I can… the mortgage market, in my opinion, is sort of nationalised right now. We have about $2.7, $2.6 trillion of mortgages held by the Federal Reserve. Another $2.6 trillion of mortgages held by the commercial banks. What’s left for anyone else is really small, maybe a trillion or a little bit above a trillion left for the rest of the whole universe of asset managers, hedge funds, everyone else. So we know that this huge increase in interest rates has resulted in significant losses just from the index, at least 10% decline in value. If we look at the balance sheet of banks, the banking system as a whole, you see the mortgage portion of their balance sheets declined by $250-ish billion while the Fed’s hiking. So the Fed must have also lost in that order because the Fed’s portfolio is about the same size in a similar mix of mortgages since they don’t report it, we don’t know it. But the market-
Bilal Hafeez (00:46:22):
Yeah, you said earlier there’s no mark-to-market so they aren’t subject to the same pressures-
Mustafa Chowdhury (00:46:26):
Losses is about $250 billion, in my opinion. So the Fed’s reps are $250 billion of the loss. Commercial banks, the way accounting works, they don’t have to take that loss in their earnings but they still have to report it. So we don’t see that having much of an effect. So $500 billion, five to $700 billion loss in the mortgage market hasn’t been reflected in anyone’s income anywhere. So it’s kind of similar, the mortgage’s nationalised, so the only corporate bonds are still held by real money in a significant way in real money portfolios and high yield as well. So I would watch out in that as the margin compressors with successive increases in interest rates, it could come from corporates, it could come from emerging markets, it could be a sudden one instead of a slowly progressing-
Bilal Hafeez (00:47:24):
And you had mentioned sovereigns because obviously sovereign debts picked up over the years and that last fiscal deaths after COVID. At the same time, when you look at the yield curve, the curve’s very flat inverted, measures of term premium. Which is the sort of extra amount that investors in theory should demand to hold longer tenors, hasn’t really picked up that much. Which you would’ve expected given the fiscal risk, inflation risk and so on. Do you think we could see a return of term premium in some way? Like a term premier shock of some kind?
Mustafa Chowdhury (00:48:00):
Right. That’s a huge puzzle for me because historically that’s always been a question. As the Fed starts hiking, we saw this conversation in 2004 or ‘05 as well, why is it that the Fed’s hiking and the yield curve is so flat? There was a constant question, I heard that same question in ’94, initially when the Fed was hiking the yield curve was not steepening as fast and the same question was raised, why isn’t it steepening? And then eventually, and it went down to sort of almost zero slope in… if you look at the two stents slope in ’94 as well as in 2004, ‘05, ‘06, the same thing. It went to 2 cents, went to about zero slope and then there’s always a different answer for why the slope was so flat. In the case of the 2004, ‘05, ‘06, eventually the story was that there was huge buying by foreign central banks, especially China and because they had these huge surpluses in their balance of payments and so that’s causing the yield curve to be so flat.
But these kind of things, when you get the answer by that time it’s kind of late and that’s when you start to see the steepening either because of a rally in the front end or because of a selloff in the long end. And in the early 2000s it was because we eventually cracked the subprime market and the front end rallied and we got massive steepening. But by that time the answer didn’t help, it’s already late and 2000 and I think ’04, ‘05 we saw the same thing that by the time we got the answer all this Orange County and all these different-
Bilal Hafeez (00:50:00):
Oh, in ’94, yeah-
Mustafa Chowdhury (00:50:00):
Players started crashing.
Bilal Hafeez (00:50:02):
Yeah.
Mustafa Chowdhury (00:50:02):
So-
Bilal Hafeez (00:50:02):
So today, I mean, what would you say? What’s the equivalent of central bank buying in the 2000s that led to a flatter curve? Is there an equivalent today of zone flow that is keeping term premier low and keeping the curve flat?
Mustafa Chowdhury (00:50:16):
It’s puzzling because it’s not as much of the central bank buying except for probably Japanese. Because that’s the only… especially Japanese private sector buying of US bonds because the Japanese yields have gone up somewhat. And so if you did the yield comparison, it was still very attractive to buy US. And so if you look at the tick data last year, that’s the only flow that seemed like it came out. But that’s not large enough to cause the US curve to be so flat. So one could be just the fundamentals this time is that we were going to get probably a coil into a hard landing and the fundamentals are that the Fed will eventually ease and ease really fast when it gets there in 2024 or at some point. Because right now market is saying there won’t be any ease anymore in 2023, which is just dramatic change-
Mustafa Chowdhury (00:51:15):
In January. But 2024, still possibility of ease, and it could be just the fundamental of the economy is being priced. Or it’s also possible that so much of the long duration is absorbed by players that haven’t been, in the past, in the market i.e., the Federal Reserve is one of them. If you just look at the Federal Reserve’s extension of this mortgage portfolio during this hike, it’s equivalent to somewhere around five, $600 billion, 10-year equivalent. I’m throwing very approximate ahead calculation but five, $600, 10-year equivalence, just the extension of duration of just the mortgage book of the Federal Reserve. So-
Bilal Hafeez (00:52:02):
Just explain what you mean by that for people who don’t understand how that would work.
Mustafa Chowdhury (00:52:06):
Because the interest rates have gone up from average mortgage just where two and a half percent, the Fed’s mortgage portfolio is 2%, two and a half percent coupon interest. The mortgage rate is now six-ish, close to 6%. So this increase of 350 to 400% will result in the homeowners, the likelihood of refinancing their mortgages extremely low. Causing the duration of the mortgages held by the Federal Reserve and held by the banking system to become way longer than it used to be. So currently if you look at Bloomberg’s estimator average duration of the mortgage index, it’s about close to seven years, 6.8 years. And it was sort of like four, four and a half before the Fed started hiking. So this change in duration for each and every mortgage held by the Federal Reserve and the banking system because between the two of them they hold almost 80% of this universe. So this changes in duration is essentially for Fed’s $500-ish billion, 10-year treasury equivalent. It’s the same as if a hedge just bought another $500 billion, 10-year treasuries without actually having to do that transaction.
Bilal Hafeez (00:53:34):
Okay, yeah, understood. Yeah.
Mustafa Chowdhury (00:53:36):
So between the two could be a trillion dollars of 10-year treasuries that they bought into their portfolios without actually buying this. But there is always another side of this. Someone else has their duration shorter like homeowners, a typical homeowner now has a shorter bond, which is very long maturity and long duration. So the typical homeowner’s ability to buy something else that’s long duration has gone up a lot. So if they were supposed to be buying a short-term bond or allocating a certain percent of their assets in short-term bonds, now they suddenly, in their own household balance sheet, have the ability to buy a longer term bond because they are short a very long duration bond to the Fed or the banking system.
Thoughts on Bank of Japan
Bilal Hafeez (00:54:36):
Okay, yeah. Now, earlier you did mention Japan and I know although you tend to focus more on the US but you have started to look a bit at Japan. What do you make of the Bank of Japan’s chances of lifting the yield curve control? So we know in December of 2022 they widened the yield curve control band to allow the 10-year JGB yield to go up towards 0.5%. The speculation now that they could remove the yield control so JBBs could then be flexible and freely floating. What are your thoughts on that?
Mustafa Chowdhury (00:55:11):
That’s a mystery, especially the governor appointee who is giving a lot of messages, everyone’s trying to figure out what he’s trying to say. My own belief is that they will try to normalise the monetary policy, it’s because it’s too artificial at this point. And then there is some realisation of whether there’s some structural issues involving the population dynamics, et cetera, that cannot be resolved just using monetary policy.
So my belief is that there will be incremental changes and whether there will be successively wider bands instead of doing away with the yield control. I also think that there was some mention of bringing the maturities for the band a little closer maybe in the five-year sector rather than ten-year. So we can see any of these happening, but I don’t believe that it’ll be the same as before and no change. Because we just started a process in December and abruptly stopped because of the change of the governor and leadership and I think that process will continue. But because of the change of leadership will do it in a way that doesn’t shock the market as much. But it will because Japan’s influence on the US risk asset market has been very strong at various times in the past and this will also be very strong.
Bilal Hafeez (00:56:50):
Now, we’ve covered-
Mustafa Chowdhury (00:56:53):
And that’s a question of timing rather than…
Bilal Hafeez (00:56:53):
We’ve covered a lot of ground, some of it technical. Hopefully our listeners have understood everything and they can obviously rewind and listen again. But I did want to ask you, Mustafa, some personal questions as well. And one I ask all of our guests is, what’s the best investment advice that you’ve ever received from anyone?
Mustafa Chowdhury (00:57:11):
I can’t point out an individual investment advice, but the one thing that’s just a common kindergarten’s things in finance are the most valuable ones. One of them is diversification. And early on I learned very deeply that since I have taken up a profession which involves fixed income and interest rates and bonds, I will invest in bonds less than a regular person who doesn’t work in the bond business does. So since the early ‘90s, late ‘80s, I’ve always had significantly higher weight on equities and less on bonds just because I wanted to diversify my career. And it worked out really well over these years. The performance from the early ‘90s to now is significantly higher for equities than bonds. And I always believe in diversifying your portfolio, and diversifying your own skills should be in that basket.
Bilal Hafeez (00:58:16):
Yeah, I mean, you speak like a true economist there thinking about yourself as a human capital invested in your career and you want to diversify against that, so that’s-
Mustafa Chowdhury (00:58:25):
Very much.
Bilal Hafeez (00:58:26):
Now, another question. Many of our audience are younger and many of them are at university, final year perhaps. Maybe they’ve just left university and they’re entering the job market, then they’re thinking about their careers and what they should do. What advice would you give to somebody who’s leaving university?
Mustafa Chowdhury (00:58:43):
The one advice that I’ve always given to young people coming into finance is to take more risk in your career, don’t take a job and be too comfortable with your job. Just look to take more risks. Start your own outfit if needed. Change jobs that are riskier but gives you more upside and don’t be afraid of taking downs, worried about downsides because things that look really negative in the short term horizon, in the long term, this all washes out. So taking more risk is my consistent advice over the years to young people that I have mentored and I still think that they should be taking more risks.
Books
Bilal Hafeez (00:59:29):
Yeah, that’s great. Excellent advice. Then another question is what books have really influenced you over your career? But it could be career related, it could be personal as well even.
Mustafa Chowdhury (00:59:39):
Lots of technical stuff that has influenced me. Things like forecasting, things like on the efficient market. I really believe in Burton Malkiel’s book on A Random Walk Down Wall Street, I really. But recently I just read this book by Alan Blinder, which I think is a master read for bonds, any investor, which is, let’s say, A Monetary and Fiscal History of the United States. And it’s a sort of reply to a book of a similar name by Milton Friedman from a long time ago.
I’ve been in the bond market since the late ‘80s, so I know what happened. At least have a decent idea about the policies, monetary and fiscal policies since that time. But prior to that, in the ‘70s, during the Boca area and prior to that, the various Fed policies and the administrations, it’s an amazingly clear description of that. And especially there’s often a discussion of two humps in interest Fed funds rate in the ‘70s and two humps in inflation. And there is a lot of different interpretations of that in terms of policy.
Blinders makes it a really clear description of the change from money supply targeting to interest rate targeting that had taken place at that time that made the inflation and interest rate, the double hump that they talk about happen. So I really think it’s a must-read for investors today and I really enjoyed it recently. Another recent book I read and it gets me thinking, it’s called the Surveillance State by Josh Chin and Liza Lin. It talks about the proliferation of face street recognition cameras and AI across the whole world.
Everywhere you go, every country, whether it’s Africa, whether it’s China, in the US, in Europe, anywhere you go there is the constant application of AI and face recognition. So where is it going? What should we be doing as individuals in society? It raises a lot of interesting questions, provides a huge amount of facts from different countries. And then I come to the conclusion maybe should we just give up and not try to have any privacy because that’s sort of futile? So this is a great book to read and I had really… it got me thinking recently this book, on the surveillance state. And both the blinders both are sort of mastery for now.
Bilal Hafeez (01:02:42):
Okay, great. No, I mean, I’ll add that to my list to read. So thanks for that. Now, finally, what’s the best way for people to follow you?
Mustafa Chowdhury (01:02:49):
I am on social media, I’m on Twitter, I am on LinkedIn, Facebook, I also can be reached on Macro Hive, so I can be reached at Macro Hive as well at mustafa.chowdhury@macrohive.com.
Bilal Hafeez (01:03:07):
Yeah, I’ll include all the links in the show notes as well. So with that, thanks a lot. It has been a great, great conversation. I learned a lot as usual and good luck for the rest of this year.
Mustafa Chowdhury (01:03:17):
Thanks for that.
Bilal Hafeez (01:03:20):
Thanks for listening to the episode. Please subscribe to the podcast show on Apple, Spotify, or wherever you listen to podcasts. Leave a five star rating, a nice comment and let other people know about the show. We’ll be very, very grateful. Finally, sign up for our free newsletter at macrohive.com/free. We’ll be back soon. So tune in then.