The corporate bond market is huge and important, allowing U.S. companies to tap investors for much needed borrowing. But even as sales of bonds have been booming in recent years thanks to low interest rates, the overall structure of the credit market and the way such debt is traded has been criticized for years. While stocks trade electronically on exchanges that provide instant and competitive quotes, a majority of corporate bond trades are still done over the phone or on platforms that tend to favor certain participants over others. Despite many efforts to improve ease of trading and price transparency in this vital market, progress has been slow. Transcripts have been lightly-edited for clarity.
Tracy Alloway:
Hello and welcome to another episode of the Odd Lots podcast. I'm Tracy Alloway. My cohost Joe Weisenthal is away, which means I get a chance to talk about one of my all-time favorite topics, which has to be bond market structure, and particularly the evolution, or sometimes lack thereof, of that structure and the way that a massive, massive market is actually traded.
So when people think about the corporate bond market, I think there's a tendency to think way back to sort of Liar's Poker-era corporate bond trading. This was the big business on Wall Street. People made lots of money from it. You had, uh, certain personalities that were tied to it, and it was sort of old fashioned the way bonds were traded and then fast forward to, you know, post-financial crisis, the 2010s and surprisingly, even as the stock market had largely electronified, the corporate bond market was still pretty much operating like it had in even the 1980s, with trades done by phone, some trades done by fax kind of amazingly.
And then if you fast forward to 2020, 2021, we've spoken a lot on the show about the idea of the pandemic forcing or increasing the rate of digitalization in the broader economy, the idea of everyone ordering stuff online and more things just moving to the cloud and computerized processes.
Interestingly, it seems like the pandemic has had a little bit of that effect on the corporate bond market as well. So we've actually seen the proportion of corporate bond trades that are electronic go up. The most recent research I saw -- I think it was from Greenwich Associates -- had electronic trading as a proportion of overall investment grade corporate bond trading at about 40% of the total. That is up from I think 30% at the beginning of 2020, and up from a minuscule one 10th of total trading in 2011. So that tells you how far we've come.
But of course there is much further to go and there have been efforts to reform corporate bond trading -- the way it's done -- for many, many years now. And one of the most interesting developments in that space has been the establishment of the fixed income market structure advisory committee by the SEC. This was created back in 2018 with the intention of improving the fixed income market and encouraging its development. So today I'm very pleased to say that we are going to be checking in on what's going on with the corporate bond market, whether or not FIMSAC has been able to fix it, or at least improve it moderately. And we're going to be doing that with the perfect person. We're going to be speaking with Larry Harris, who was on FIMSAC, so on the SEC committee, he's also the former chief economist at the SEC -- he was there from 2002 to 2004 and he is currently the Fred V. Keenan Chair in Finance at the USC Marshall School of Business. So Larry, thanks so much for coming on the show.
Larry Harris:
Oh, it's a pleasure to be here.
Tracy:
So maybe just a first question. I mean, it sounds like you have been in the trading and market microstructure space for a very long time. Is this something that you've always been following?
Larry:
Yes. I've been working in market micro structure pretty much from the beginning of the academic field now probably 40 years.
Tracy:
So given your expertise, I have to ask what is it about corporate bonds that seems to make them a little bit more stubborn when it comes to things like electronic trading, standardization, other efforts to improve liquidity? It feels like the corporate bond space, despite people trying to evolve it for, I mean, decades now, it feels like it's definitely an uphill battle.
Larry:
Tracy, that's a great question. If you listen to the vested interests in the bond market, they'll tell you that bonds are just simply different. They're different from equities, they're different from options, they're different from futures. And they say that the differences are what makes the market structure different. Of course they don't often articulate exactly what those differences are. So often they'll tell you that they're different because there are so many bonds, but there are even more options that trade and they trade in exchange markets and they trade pretty well. They'll tell you that they're different because they're so risky. Although equities are more risky per dollar of principle than our bonds. So it's a little hard to understand this question on its face.
We're going to have to dig deeper, but as long as we're starting with some history … You started around the eighties, I'd like to push it back much further, to share with you and our audience some very interesting things that many people don't know>
Tracy:
Sure!
Larry:
Bonds used to trade almost exclusively in exchange markets — at the New York Stock Exchange and at the American Stock Exchange, corporate bonds stopped trading on the exchange markets in the mid forties. And an interesting paper was done by two academics, Bruno Biais and Rick Green. And what they did is they look to see how those bonds traded in the forties and in the late thirties and compared that trading to how presently trade now. And when I say how they trade, I'm really referring to how expensive was it to trade the bonds. And when I speak about expense, I'm not speaking about the price I'm talking about the transaction costs of trading. So if you buy a bond that's worth a hundred, but you pay 101, then your cost was a one point of par value.
So these bonds that traded on the New York stock exchange, I'll save the result for a moment, but you can all anticipate what it's going to be. Now, let's talk about how they traded. So they traded in order books, just like stocks trade, but because there were so many bonds, they had to put the order books in filing cabinets. And so these order books, these databases were contained in filing cabinets. If you wanted to trade a bond, you went through the bond specialist post and you mentioned, you know, you said, what bond you wanted to trade. And the clerk would go, the specialist would go and look up the bond in the filing cabinet, find out who had left orders for it. And so now the punchline of course, is one that everybody's expecting. In those non-electronic, but order-driven markets -- so order-driven means exchange markets in which you have rules that match buyers to sellers.
So the most aggressive buyer, the one is willing to pay the most gets matched to the most aggressive seller, or the one who's offering the lowest price. In those order-driven markets, bonds traded at lower transaction costs ages ago, then they were trading until just a few years ago.
So that leaves us with the interesting question: why? Why did the markets move away from the stock exchange or the stock exchanges, because the AMX bonds were also important. Why did they move away from the stock exchanges and into the investment banks and the dealers and brokers? So somehow they captured it and this is not really well understood. Bruno Biais and Rick Green speculate about it. It's hard to imagine that it made sense for the buy-side, the investors, because in the end they did worse. But, I think it's just because the investment banks probably controlled much of the supply of bonds. They were buying and selling them. They were dealers and they preferred to trade as dealers than as dealers operating in exchange markets.
Of course dealers always operate and exchange markets, there's nothing new there, but they're more powerful when they trade in a dark environment where you don't see the trade prices and you don't see the quotes. Now fast forward in 2002 to ’04 or so. At the time I was chief economist of the SEC. FINRA started the TRACE bond price reporting project. And the dealers were very opposed to having those trade prices reported even with a 15 minute lag, which is what was finally determined. I was privileged along with some others to engage in research that showed that making those bond prices public information, would lower investor transaction costs by about a billion dollars per year.
So the opposition to the dealers to making the prices public kind of withered away in the face of those empirical results. I'll tell you two fun stories, they’re somewhat self-serving, but it does make the point about this story.
Tracy:
Go for it.
Larry:
Yeah. So the first one is that, so the SEC did indeed mandate that the bond trade prices be made public and now TRACE prices are available on the internet and everybody can see them. So the two stories are first, somebody came along, I forget the fellow's name, might've been a woman, and replicated the study in which I participated and discovered that indeed investors were saving a billion dollars. So my study had been based on just a trial of TRACE and from the bonds that were public, and we compared them to the bonds that weren't public, where the information wasn't publicly available – the trade price information. Doing the comparison, we were able to determine what we thought the savings would be.
The way we did the study we compared how the bonds that were trading in more transparent markets, the initial TRACE transparent bonds, we compared how they traded to the ones that had not yet been converted. And we found that about a billion dollars would be saved by investors. And so the first story is that indeed somebody looked at it a few years later and determined that our estimate was on target. So I say somewhat immodestly that we did well, but the important point is that not only on a prospective basis, but on a retrospective basis, it appears that that bond price transparency substantially lowered transaction costs. So that's the first story. The second story is a little bit more fun. Anette Nazareth who was then the director of what was then called the Division of Trading and Markets, who was a big proponent of making the bond prices transparent.
She kept asking me, Larry, how come you're not done with that study? How come that you're not done with this study? Because they needed that study to build the political wherewithal to get the capital to get this done. And I kept telling her, listen, Annette, I'd like to have it done yesterday, but we're going to be super careful because the dealers are going to criticize this study. So fast forward about 10 years, I’m at some conference, I'm talking to somebody, you know, an academic conference. And this guy sorta drops, he says, you know, you did that bond study about transaction costs and the initial TRACE data. And I said, yeah, he says, I know a backstory about that. I’m going, yeah. He says the dealers sent it to a colleague of mine at a university — I won't say which one, but it was a strong reputable university — asking is there anything wrong with the study? You know, can you find the econometric problems or something like that?
Tracy:
Wow.
Larry:
And the response was apparently ‘no, you're going to have to live with it.’ It was well done. So again, a pat on my back, but also there's a more important point, which is that there's a role for academic work, rigorous work, doesn’t have to be done by academics. It can be done by practitioners as well. And by the economists at the SEC. There's a role for rigorous work for trying to understand how things could be different.
Tracy:
That is such a fascinating insight into the way the SEC works and also the way its various stakeholders sort of interact with it. There's one thing I wanted to ask you before we sort of move on to what happened next, but can we just, can we focus in on the dealers here, a little bit more. So, you know, you, you kind of touched on this, this idea that the dealers have a vested interest in the way that the current system works. So I brought up Liar's Poker at the beginning. You know, this is the idea of people sitting at the big dealer banks who are taking calls from investors who want to buy or sell certain bonds. And they're basically the middleman who are holding the informational power and who can say, well, I have a quote on this one or I have a quote on that one and no one really is quite clear what the actual price of these bonds are. They're sitting in the middle. It's a very intransparent process.
Now in the years since then we have had these new electronic trading venues set up and here I have to do a massive disclosure and say that Bloomberg has one, so we have Bloomberg, TradeWeb and MarketAxess. And my understanding of those is that basically they've made dealing with dealers potentially more efficient. So you can do a request for a quote with just, you know, a click of a button and you can get one very, very quickly, but they still preserve some of the power of the dealers in that process. You can't do a deal directly with another buy-side participant, for instance. Is that right? Is that the right way to think about it?
Larry:
Yes, that's generally correct. And though they would like to create exchange type markets, they know that they still need the dealers to provide the liquidity. And so the, the dealers call the shots. Tracy, you use the word informational power. You could not be more on target with that phrase. So, just so that everybody listening has a clear understanding of how powerful information is, let's discuss a simple example that everybody can relate to. Suppose you need to buy, say a used car, or it could be a new car. And you go to the dealer and you find what you want to buy and you then make a bid. Okay. Now there's two mistakes that you can make, you can underbid, or you can overbid. Now if you underbid here's what's happens. So the salesman says, say, Larry, I'd really love to sell you that car at that price. But if we do that, we just can't stay in business. The price is simply too low.
So you don't do the deal. Now suppose that you overbid, what does the salesman say? Salesman says, Hey, Larry, I'd really like to sell that car, but if I do so at that price who we're going to go out of business. But hang on a minute, let me see what my sales manager says. And of course he comes back and sells you the car at the price that's too high. So what's going on here? Obviously you can make two mistakes, but because the dealer knows values better than you do, the dealer isn't gonna sell it for undervalue, but it'd be more than happy to sell it for overvalue. And so this is the story. This is why information about values is so incredibly important.
Unfortunately, we only have half the story on price values. We've got now the TRACE data that we just talked about, which tells you the prices at which bonds traded by and large. We don't have very much quotation data. Although it's increasing substantially, we don't have much quotation data where the dealers are telling the public the prices at which they're willing to trade. So if you want to go buy a bond and you're smart, what you do is you go and call up a dealer and say, what's your offer? And then you call another dealer and ask for their offer and you call another dealer and you sorta make them all know that you're doing this. Or you can do a request for quote, which is essentially the same thing. And you try to get them to compete with each other, but it's an expensive process. And they're only volunteering information to you, not to the public. So it's not quite as effective as the competition. We see every single day for even illiquid securities at the various stock exchanges or the options exchanges.
Tracy:
So when pricing data became available on TRACE, you're getting the specific points, the specific prices at which bonds are being traded, but you're not getting potential quotes from a bunch of dealers about where they could buy or sell them. You're getting the actual transaction data. How did that change the market or change behavior of participants?
Larry:
Well, sophisticated people and even some less sophisticated who know enough to look up bond prices can find out where the bonds recently traded. And if a bond recently traded, you know, just minutes ago, then you have a good point of reference to what the value of that bond is. Especially if it's a very large trade, the large trades tend to be more informative than the smaller trades because the smaller traders, frankly, they just don't know as much. They don't know how to control the information environment. They trade through brokers who aren't particularly helpful. And so the smaller trade prices aren't so important. But if you want to trade a bond, say you're trying to sell a bond. That’s a more typical story than wanting to buy one. You want to sell a bond that you need to generate some liquidity.
Maybe that bond didn't trade yesterday or today, that would be ideal. Say it traded last, you know, a month ago or perhaps a year ago. There are a lot of bond issues out there and some of them, we say they trade only by appointment, which is kind of funny. Okay. So you got a bond trade from ages ago. And now the question is what's the thing worth? And you know, there are bond pricing services that can give you some sense of what the bond is worth. That's very important because mutual funds that hold bonds have to value their bonds. And if they don't trade well, I mean, what's the bond worth? They’ve got to know every day. So these pricing services, what they do is they say, okay, this bond didn't trade, but it looks a lot like these other bonds, maybe bonds by the same issuer, or maybe bonds with the same coupon rate, the same maturity, the same preference in liquidation, stuff like that.
And so they say, okay, we got a model that says, if these other bonds are sort of recently traded and we see their prices, we can infer what this thing is probably worth. Of course, you know, the small trader, that guy who’s trading odd lots...
Tracy:
You’re on the right show.
Larry:
I know, that's why I laughed. Yeah, they don't get that information and it's expensive In any event though, what the pricing services say is not necessarily true either. It's just their best estimate. Real truth comes when people are ready to put their money on the table and trade and actually negotiate prices. So it's good to have those TRACE prices, but it would be even better if we had prospective prices, pre-trade prices. The quotes.
There are a lot of them out now that are out there now. And entities like BondCliQ are starting to aggregate them from dealers. The dealers are willing to share with BondCliQ because they want to see what the other dealers are saying. Dealers are, they're not not dummies. They're very, very sharp people, of course. But the consequence though is that increasingly this data is becoming more public and, you know, BondCliQ appears to be leading here, but I suspect, I believe that there are some others as well. The dealers aren't entirely happy about it. So it's not an easy proposition.
As long as we're speaking about a little bit of history, Cantor Fitzgerald, ran an open outcry, sort of brokered bond market in Treasuries and through their E-Trade subsidiary, since spun out, they wanted to get the bonds into an electronic trading system. And they had a terribly difficult time doing that. The dealers basically boycotted the system for awhile, but they persevered. They managed to get it done, but it's not easy.
The big challenge that these systems face in creating an exchange-like system is that when somebody needs to trade, they usually need to trade right now. And they look at a system, they say, wow, that is a Cracker Jack system. That's beautiful. I'd love to trade there. But right now I got to get my order done and there's nobody else there. So I'm going to get my order done. And when that market becomes liquid, you give me a call and I'll definitely send my orders there. So it's a chicken and egg problem here. Only the problem is not which came first, it's that neither comes first because people need to get their business done. And so that's the force that keeps change from happening too quickly.
But on the flip side, these electronic systems have vastly decreased the cost of dealing. And now you have lots of proprietary traders who are willing to deal bonds. Like they deal stocks, they deal odd lots, but they're willing to come back over and over again. And so increasingly you're going to start seeing -- we've already seen it -- algorithms used by institutions to fill large orders in odd lots. And they'll do that because there are dealers who are, well, dealers are proprietary traders, but let's call them proprietary traders, essentially using, you know, some crude version of the high-frequency trading methods that they use in the stock markets to deal.
We should be clear about what bonds are. So here's a fun way of thinking about a bond. People tell you that bonds are something special, but from this point of view, they're perhaps not so special. Bonds, if you think about their risk, they're a package of two types of risk. There's interest rate risk, and it's bundled with credit risk. So the credit risk is the risk that the company that's promised to pay off its bonds goes bankrupt and you don't get paid. And interest rate risk is the risk that say you bought a bond and then interest rates rise and the price of your bond drops. So that's not comfortable.
So think about this risk. What is there about putting both risks into a single instrument and tying them together with an invisible string? The string is actually called the bond covenant, the contract that sets up the bond. What is there about doing that, that makes the bond something different from the risk that trades elsewhere? So the interest rate risk trades in highly liquid Treasury bond markets, they're are also in the futures markets.
Whether they are cash or in the futures, they trade in these order-driven exchange-like systems and the credit risk, as we mentioned earlier, the credit risk of bonds is as much less than the credit risk of the associated stocks. So the credit risk is being traded in, again, order-driven markets, the stock exchanges and the similar systems. And they're both highly liquid markets with lots and lots of pre-trade and post-trade order exposure. So pre-trade is the quotes and post-trade are the trade prices and those markets function really well.
So what is there about that magic string that connects those two risks into a single bond that somehow makes it necessary to trade it in the dark? And if you've got a good answer, I'd love to hear it because I've heard lots of people try to explain it, but it nobody's convinced me. And I'll remind you again, one final reminder, before we turn to another topic, that explanation, the magical explanation, also has to explain why it didn't operate in the 1940s when these markets were trading in ex-ante and ex-post transparent markets at the exchanges. So anybody who wants to explain this one is going to have to work really hard to be very creative.
Tracy:
Well, I think that brings us up to modern times, very, very well. So, you know, we have some progress when it comes to pricing transparency, notably the TRACE data, which is now being produced based on actual bond trades, but not potential quotes. You have the electronic trading venues that are mostly operating on a request for quote basis and are mostly preserving some of the informational power of the dealers. And you have some venues that are trying to aggregate prices and quotes for bond trades, like BondCliQ, which you mentioned, but sort of improvements on the edges, nothing massively fundamental. And yet this is something that, you know, you laid it out very clearly earlier. This is something that has been of concern to the SEC for a long time now. And I personally remember speaking to, I think it was Dan Gallagher back in 2014 and he was saying, you know, it's time to address the Liar's Poker dynamics embedded in the corporate bond market system.
And yet here we are years later still talking about potential fixes or improvements that could be made. So I want to talk to you specifically about one of the big SEC projects to try to fix all of this, which is the establishment of FIMSAC, the Fixed Income Market Structure Advisory Committee in 2018. My understanding is it was supposed to be the big effort by the SEC, to try to come to a consensus on what needs to be done and then actually implement it. And you were on the committee. So maybe you could give us your perspective of what exactly the goals were when it was established?
Larry:
Well, I suppose it depends who you ask and how candid they will be when they respond. When I went into it, I thought this was a great opportunity to modernize the markets. I had been instrumental with Anette Nazareth and many others. Remember every success has lots and lots of grandparents and parents and godfathers and godmothers and so forth. We had been instrumental in moving the stock markets from open outcry floor-based systems into electronic systems that are incredibly efficient now. So that was neat. And so I thought, gee, this would be a really great time to see what we can do to, to solve some of these problems. They're not really problems, but just to make the markets better. And so some things that I was hoping we'd be able to do is we could, you know, call for the formation of a national best bid oroffer or have Finra do like Nasdaq did … Create a Nasdaq, the original Nasdaq for bonds. The original Nasdaq was just a quotation system where you showed everybody's quotes and all the dealer quotes, it's called a quote tableau. So just seeing the quotes ex-ante gets people to compete more and makes the markets better. And then they ultimately evolved into the Nasdaq stock market, which is a full-blown exchange and unregulated as such now. So I thought, you know, at a minimum we could try to ask for that. There was almost no support for that. There were a few people who were in the boat, but the vast majority of the members of FIMSAC were opposed to any change so radical.
Tracy:
Who were the other members?
Larry:
Well, we'll talk about that. But so we weren't calling, you know, this proposal was not a call for, you know, a trade-through rule, which would be pretty radical, we were just saying, you know, let's collect all the ex-ante data, the quote data. So who were the people on FIMSAC? Well, by law, these advisory committees have to be representative of sort of everybody who has an interest in the issue. Everybody has to be at the table, but the law doesn't tell you how the table should be weighted. You know, how many people of each type. And so you had the bond dealers, you had the bond issuers, you had a few academics, you had a few investor advocates. You had the alternative trading systems. You had some dealers. I think I mentioned them as well, issuers and large institutional investors.
So I think it pretty well covers the gamut, but it was weighted heavily towards people who were either vested in the status quo or who were dependent on people who are vested in the status quo. So if you're an issuer, you don't want to go out and off your investment bank because you’ve got to do business with them. And so basically you just go along and, you know, I would make impassioned speeches about how we should be doing what's best for the country and stuff like that. And they all in their own way thought that they were doing what's best for the country. But I think that perhaps their points of view might have been partially compromised or at least, subtly influenced by their self-interest. So I'm trying to be polite, but you know exactly what I said. And it's not surprising. There's a lot of fear that if we made any changes, we'd screw things up. And so the problem to getting change is that our bond markets are for all their warts, the best bond markets in the world. They could be a whole lot better. I know that from my experience and a lot of other people know it, but getting there is really difficult. And so it was very challenging — and disappointing.
Tracy:
My understanding was, I mean, this can't have come as a surprise to anyone who created this committee. Like if you get a bunch of varying interests together in the room, they might not be able to agree on fixing the market or improving the market, especially given that they haven't really agreed on how to do that for many, many years previously. And I always thought the idea of getting the SEC involved was that you would have an independent arbiter who could look at this information and try to come up with a solution that might be independent of the individual interests of people in the market currently. But that doesn't seem to have happened judging by the lack of news that I've seen on, you know, stuff being proposed or coming out of the committee.
Larry:
Well, there's a lot in that. So I mentioned that the composition of FIMSAC basically established what they were going to do and what they were not willing to do. It's instructive that the very first thing that they proposed was to delay the reporting of super large bond trades. Right now, prices are reported within 15 minutes with a marker saying that the trade is, if it's an investment grade bond, it's more than $5 million. And then six months later, you can find out the full size, but who cares at that point? And $5 million trade is plenty big enough, you know, it's big. So they propose that we should delay reporting those trade prices and sizes for, I think, three days or something like that. But then we'll report the whole thing. So it was like, well, we'll give you something. But what they were taking away was really, really meaningful.
And this was the first thing they proposed. So that was like a shock to me. I was just like, we're going backwards. It’s just crazy. Okay. So the SEC appoints all these people and they, according to how they do the weighting, they can determine the outcome. So let's be generous to the SEC and say that there are a couple of different ways that change can take place. And how you feel about these ways depends on how you feel about uninformed investors and how you feel about political economy or your political philosophy. So one way is for the SEC to mandate a change. So we did this when we were making changes to equity market structure to make the markets electronic, that was Reg NMS and it was extraordinarily successful. So the SEC has the power to regulate markets and it gets challenged by the incumbents.
But in the end, if they do their work properly, they can get what they want done. The argument for the SEC doing this is that somebody has to represent the interest of uninformed investors who will never be able to do this by themselves. Okay. So that's an argument. The counter argument, which is also respectful, is that listen, if the markets would be better in a changed state, there will be people who will provide those changes. And though it might be difficult, it's not impossible. And unless it's impossible, we shouldn't regulate. So for instance, there are private entities who are now aggregating best bids and offers. We talked about BondCliQ, other brokers are doing it as well. So Interactive Brokers will allow its clients to trade on bond markets where Interactive has collected quotes from a multitude of dealers. And more interestingly, Interactive will allow their clients to post their offers to trade.
So they can post bids and offers just like a can in the stock market and, they post them at venues where they actually might get hit. And so instead of always buying at the ask price, you might be able to buy at the bid price, if you trade through Interactive. Now in the interest of full disclosure, I have to tell everybody that I am a director, I'm actually the lead independent director of Interactive. And so you should recognize that that I have an interest in letting people know about this, but they're innovative. And that's a neat thing. So going back to political philosophy, clients want this, they can go trade through Interactive Brokers. And if Schwab sees that they're losing clients or ETrade or Ameritrade, they can start offering the same services. So the premise here of course, is that ultimately the uninformed and an unknowledgeable trader will benefit because there are knowledgeable traders who put significant demands on the system and cause the system to change.
That may take 20 or 30 years or may never happen. It may only happen for the informed traders and everybody else sorta gets left behind because people never know any better. And if that bothers you, then you're back into the first camp that says maybe the SEC should do something. So from my point of view, I think that the SEC should exercise a little bit more power. I don't like to see them do too much regulation, but they should exercise power when you have certain problems in the market that make competitive solutions difficult. And those problems are agency problems where people are represented by people who have conflicts of interest. The people who are doing the representation, like the brokers, they don't have a strong interest in serving their clients well, if the client doesn't know that they're getting screwed, I believe that's the technical term in the market. So hope nobody takes any offense.
Tracy:
It wouldn't be the first time it was said on all Odd Lots, I can assure you.
Larry:
Yeah, I'll be more careful about some of the other technical terms that are often bandied about in the markets. So when the broker has a conflict of interest and the client doesn't even recognize it, the broker is not going to be too eager to change it because a broker's going to act in their self-interest and their self-interest is usually to go along and keep things simple. And in some cases, they even get paid for order flow. But I don't think that happens in the bond markets. I may be misinformed. So that's a potential problem. And then another potential problem with this, economists call it the order flow problem. But if you're just a regular person, you just know it as a notion that liquidity attracts liquidity. And we discussed it earlier. When we talked about how difficult it is to start a new market, so you got a new market…
Tracy:
Right, the chicken and egg problem.
Larry:
Exactly, it’s a Cracker Jack market, but the problem is that it never gets off the ground until people are willing to trade there and they're not willing to trade there cause they got to get their stuff done right away. So that's it, that's a problem that inhibits competition. We all may be better off with that Cracker Jack market. You know, maybe it's an order-driven market with all the bells and whistles that will make trading super efficient like we see in the equity markets and then to a lesser extent in the options markets and certainly in the futures markets. We may want that and it could be better, but we may never get there without the assistance of a regulator. So these problems, agency problems and externalities are the fancy words that economists know and use, these problems ensure that free markets don't always produce competitive markets. And so this is where political philosophy he can diverge. I am a hundred percent in favor of free markets when free markets produce competitive markets. But if a free market has got problems like agency problems or externalities that ensure that we don't get to the competitive solution, then we need to have a very light hand that will give a nudge. I'm avoiding the word, that nasty word regulation, but to give a nudge, to get us to the right equilibrium, to get a structure that benefits everybody.
We should talk for a moment why we care about this. Even though we have the very best markets in the world in the bonds, they could still be substantially better. And if they were better, we would all benefit. So volumes would increase substantially, which ironically would benefit the dealers, transaction costs would drop. And we would, you know, people who are saving for their retirement would be able to save more efficiently and issuers would see lower issuance costs because the cost of issuing a bond depends on how liquid the market's going to be after you've issued the bond. People don't want to pay so much for a bond that will be locked up forever. If they need the money out, they don't want to lose a lot trying to sell it.
But if the bond looks like it's going to trade in a highly liquid market, they'll pay more for the bond. And when people pay more for a initial public offerings of bonds, when the corporations are funding it, it means their funding costs are lower. And so that's good for them as well. So there are a lot of really important and very valuable benefits that are associated with making these markets better markets, but there are some strong vested interests. I’ll give you a story about vested interests. So the SEC adopted TRACE, we talked about it at length. At the same time, the Canadian markets were faced with the same proposal and they didn't do it. They since have adopted post-trade transparency, you know, to show you what the trade prices were, but it took them 10, 15 years to do it. And the reason perhaps was that the regulator there was the Bank of Canada and the Bank of Canada has a close relationship with the large dealers. And as a consequence, they just didn't want to rock the boat. This despite the fact that there's overwhelming evidence from America, that the world didn't end, as some people suggested, with those bond prices being made public. And likewise, the world's not going to end if bonds were traded in order-driven systems. Because we see order driven systems trading, similar instruments all the time, all throughout the world and in the United States as well. So lots of fears about the end of the world, but I'm not there.
Tracy:
Well, can I just ask one devil's advocate point on this idea of regulation versus natural market evolution? And it kind of relates back to something I said in the intro, which is this idea that over the past year or so you have seen something of an improvement or a migration in the amount of bond trades that are actually electronic and going through some of those platforms. So I think I cited the Greenwich Associates data that had the number of IG — investment grade — bond transactions that were electronic at 40% currently, versus something like less than 30% at the beginning of 2020. So something happened in 2020, you know, the pandemic, you had the Federal Reserve buying corporate bond ETFs for the first time. Those two things arguably have helped shift the market more than previous regulation or some of these previous private efforts have done in the past. So is there an argument to be made that you could just let the bond market continue its natural progression and maybe eventually people will change their behavior?
Larry:
Certainly you could make that argument. There's strong evidence that this is what's happening and it may continue to accelerate. The two stories that you offered may have contributed. But I think there was another story even stronger, which was that as a result of Basel III, the large banks came to be at a disadvantage to proprietary traders when dealing bonds. And we can talk about what that disadvantage was but basically when holding a bond portfolio, the banks have to hold more capital against the positions than proprietary traders have to hold. And as a consequence, the banks could not compete as effectively as they had competed before, which meant that a lot of the personnel at those banks left and formed their own proprietary trading groups or joined other proprietary trading groups. And all of a sudden you have a core of highly skilled dealers in a proprietary environment that want to make money.
And they're electronically sophisticated. And I believe that it was probably these guys who have made the electronic venues more liquid and has given us that 40%. And I expect that trend will continue. And I think that's, I think that's a good thing. And indeed, if you look at studies of spreads, we see that spreads have indeed dropped in the bond markets because of more competition from these electronic entities. And there's of course two competitions here, we want to promote both of them. There's the competition for best price, which works best when you put everybody in the same place. Best price means a buyer looking for the lowest purchase price and the seller is looking for the highest sales price. And you also have the competition to be the venue that hosts that competition, whether it's an exchange or a broker or just a dealer who says, I'll take the other side and give you the liquidity.
So we like to have both competitions, but the two competitions don't co-exist particularly well with each other and the SEC over time has leaned towards promoting one competition versus the other competition. And what's really interesting of course, is that anybody who has a position in these markets — and by position, I don't mean their bond position, but I mean their opinion about how they should be organized — anybody who has such an opinion always cites their opinion as being pro-competitive. So they want to see more centralization. They say they want to see the competition for best price enhanced. That's really good. And if they want to see more competition among exchange venues and so forth, they say the system as it is, is really competitive and that's important. And it's created the best markets there ever were.
Until recently before these electronic innovations, those markets back in the 1940s at the stock exchange, they were better. And how much better would they be now that we have computers that can handle those filing cabinets? You know, it’s just a database problem. So it's pretty darn interesting. One other issue I’d love to share with you, the SEC’s regulatory framework for small retail traders in the bond markets, is dealer centric. The assumption is that dealers are doing all the trades. In contrast in the equity markets, the regulatory framework is broker-centric. It revolves around the brokers. And so the difference is that the brokers are required to get best execution. But when the brokers are acting as dealers to you, there's a sort of a different relationship, but quite frequently, now that the markets are electronic and where you have venues where quotes are being aggregated, or we have dealers who are sharing their quotes directly with brokers, quite frequently, a broker dealer who offers bonds to their clients takes no position whatsoever.
They do what's called a riskless principal trade, which means that they buy from a dealer and they immediately sell, usually marking it up to their client. And they'll show the client what's available because they have a list with what the dealer says available in the client, chooses what they want and bingo the trade is done. Okay. So there's no principal risk involved with that transaction because the intermediary, the broker is literally acting as a broker, but is regulated as a dealer. So there's a type mismatch here. And if there's a good reason that we regulate broker dealers, primarily as brokers, when doing riskless principle trades, which is essentially what they do when clients buy stocks or options or futures, why don't we regulate the same way when they're trading bonds? And the answer of course is bonds are different. But I don't think they really are so different.
Tracy:
There's one other question that I wanted to ask you, which is about Gary Gensler, the new SEC commissioner, and, you know, we've seen him come in and he's made a lot of noise about focusing on things like crypto and payment for order flow and the equity market, and you know, general retail stock trading issues seem to be high on his agenda. But he has also spoken quite a bit about price transparency in the bond market and maybe some other improvements that you could make to the regulatory structure. And I'm aware that we haven't really spoken about ATS at all, or alternative trading systems, but is this something that you would expect Gensler to be looking at in the next year or so, do you get the sense that this is high on the SEC’s priority list given what happened with FIMSAC?
Larry:
All that happened under FIMSAC and most importantly, the constitution of the committee itself, took place under an SEC that was dominated by more conservative political interests. So now we have an SEC that’s now dominated by a somewhat more liberal political interests. And as a consequence there is a potential for a change here. My impression though is that this is not his hot issue. There's a variety of reasons why, but the evidence that it's not a hot issue is that he still has not appointed a full-time, Director for the Division of Trading. Its’s now called the Division of Trading and Markets. It used to be called the Division of Market Regulation. I misspoke earlier when I mentioned that. So he still has an acting director who seems quite competent, but an acting director just doesn't have the same power as a director.
And then all of his regulatory people report to a person in his office who doesn't have a strong background in market structure. She's an attorney. Her last job was a deputy general counsel for AFL-CIO. Uh, I've never met her. I'm sure she's a wonderful woman. And I'm sure she's a very fast learner as almost all attorneys are. And I can assure you, every attorney thinks they are, but it's not the same thing. And so just from those omissions, if you will, I'd suggest that perhaps market structure issues aren't high on his regulatory agenda.
Now, all that said, there's a lot of money involved in these issues. And when there's a lot of money involved in the issues vested interests will lobby their senators. And so they, they make contributions to senators and senators look at that and say, Hey, the U.S. bond markets are the best markets in the world. Why would we ever want to muck with them? Just ‘cause some academic who thinks he knows something tells us that it could be better. Well, you know what, I just don't see it. And so, you know, if the SEC proposes to do something, a senator writes a letter saying — not in opposition — but can you kindly explain why you're doing what you're doing? And then in between the lines, which is not written, it’s oh by the way, you may recognize that I sit on the finance committee and we set your budget. And from the tone of letter, there's a little bit of skepticism and sort of everybody knows what's what, so why do senators do this? They do care about the markets, but it's an abstraction that’s far away from them. They can raise capital, which is political contributions, cheaply, and then spend it where it's more dear to them on the issues that are more important to them, whether it be abortion or early childhood education or armed forces, or who knows what — roads — doesn't matter.
And so what you see is senators on both sides have this tendency to be co-opted by a strong interest. And so what you need is a very, very strong SEC that can make the case and explain how much better things would be if we do all this, because it's going to be painful. The SEC will end up being sued, there'll be letters from senators. And at some point the letters can be less than subtle. So you need somebody who's really been empowered. We had that under Harvey Pitt, who was actually too outspoken, but it continued, under Donaldson. And that's when Reg NMS was adopted over the objections of the Republicans. And it was odd that Donaldson, a Republican, actually voted with the two Democrats to adopt Reg NMS. But that was perhaps because he was closer to the end of his career than to the beginning of his career and may have been thinking more as a statesman than as a, somebody with vested interests. He also may not have cared very much and just let staff do what they wanted to do, and they got away with it. But it's difficult. No question about it.
Tracy:
Yeah, I'm getting that sense. Larry, this is been a fascinating conversation and I feel like we could probably talk all day potentially about SEC history and some of the political machinations there, but thank you so much, really appreciate you coming on Odd Lots.
Larry:
Thanks for this opportunity to share some really important insights with you and your audience. Okay. Take care.
Tracy:
So clearly I enjoyed that conversation. It's always a joy for me to get back to talking about corporate bond market structure. And of course, one of the things that stand out is that tension between a regulatory push towards fixing these agency problems that Larry described versus the natural development of the market. And part of me thinks, yes, we've seen some improvement over the past couple of years or so, but there's still so much further to go in what is one of the most important markets in the world. And it's sort of amazing to me that there isn't more of a spotlight shone on this particular issue. But on the other hand, I thought Larry did a very good job of explaining some of the political considerations that go into formulating the SEC’s sort of agenda under various personnel and new commissioners.
And so maybe that explains it. I can't imagine that telling people that your regulatory focus really needs to be on fixing credit markets is that much of a sexy topic for a broader audience — although, you know, certainly on this particular podcast, we try to make it one.
All right. I think I'm going to leave it there cause it's weird to talk to myself without Joe.