Liquidity and Trading in Fixed Income: Why Bonds Are Harder to Trade Than Stocks
You can build the most brilliant bond portfolio model in the world. But if you can’t actually trade the bonds you want, none of it matters.
That’s the core message of Chapter 9. Richardson shifts from the theory of portfolio construction to the messy reality of getting trades done in bond markets. And “messy” is the right word. Bond trading is fundamentally harder than stock trading, and it’s not even close.
Bonds vs. Stocks: The Liquidity Gap
Here’s a quick comparison that puts things in perspective. As of 2020, US Treasury bonds, corporate bonds, and stocks all had market capitalizations above $10 trillion. Similar size. But the trading activity is wildly different.
US Treasuries see about $600 billion in daily volume. US stocks see about $500 billion. US corporate bonds? About $40 billion. That’s it.
The number of securities tells a similar story but in reverse. There are roughly 100 Treasury securities and about 4,000 stocks. Corporate bonds? Around 10,000. More things to trade, way less trading happening.
Bid-ask spreads show the cost difference too. Treasuries cost less than 0.05% to trade. Stocks cost less than 0.1%. Corporate bonds cost 0.3% to 0.5%. When you adjust those trading costs for the volatility of each asset class, corporate bonds are the most expensive to trade by a wide margin.
So right out of the gate, anyone investing in corporate bonds faces a structural disadvantage in execution.
The Over-the-Counter Problem
Stocks trade on exchanges. You see prices, you click buy, you’re done. Corporate bonds don’t work that way.
Corporate bonds trade “over the counter.” That means you need a dealer. Dealers act as middlemen, holding inventory of bonds and quoting prices to buyers and sellers. Think of it like buying a used car. There’s no central lot with posted prices. You call around to different dealers, ask what they have, and negotiate.
This creates real differences in how bonds are quoted and traded. Investment-grade bonds are typically quoted on a spread basis, meaning dealers give you the yield difference over the nearest Treasury bond. High-yield bonds are quoted on a price basis. In both cases, the quotes are indications, not commitments. A dealer might show interest in trading a bond, but when you actually try to execute, the price might move or the quantity might not be available.
The difference between liquid and illiquid bonds within corporate credit is enormous. A bond from GE might have a dozen dealers competing with tight spreads and large sizes. A bond from a smaller issuer might have a handful of stale quotes, wide spreads, and tiny sizes. Same asset class, completely different trading experience.
The Disappearance of Dealer Inventory
Here’s one of the most striking facts in this chapter.
At the end of 2007, primary dealers held nearly $265 billion in investment-grade corporate bonds. By the end of 2013, that number had dropped to $73 billion. By the end of 2020, it was just over $4 billion.
Read that again. Dealer inventory went from $265 billion to $4 billion in about 13 years.
Why? Regulatory changes after the financial crisis made it more expensive for banks to hold bond inventory on their balance sheets. So they stopped. Dealers went from being principals (holding inventory and taking risk) to being agents (just matching buyers and sellers).
This matters for everyone trading bonds. There’s less cushion in the system. When you want to sell, there’s no guaranteed buyer sitting on a pile of inventory. You need someone on the other side of the trade, and finding that someone takes time and effort.
Credit Indices and ETFs: The Liquid Alternatives
If single-name bonds are hard to trade, what about getting credit exposure through other instruments?
CDX contracts (credit default swap indices for North America) and iTraxx (for Europe) offer an alternative. These are standardized derivatives that trade on swap execution facilities. The liquidity is excellent. Bid-ask spreads are less than 1 basis point. You can trade large sizes within minutes. The catch is tracking error. CDX and iTraxx don’t perfectly match corporate bond indices because of differences in seniority, maturity, sector composition, and issuer selection.
Bond ETFs like JNK (high yield) and LQD (investment grade) trade like stocks with tight spreads and deep order books. But they come with management fees, and their returns tend to lag the underlying indices. The create-and-redeem process that keeps ETFs in line with their underlying bonds also introduces the possibility of adverse selection.
Total return swaps give the closest match to the actual index but come with high costs and counterparty risk.
Each tool has its place. Richardson notes that credit index derivatives are particularly useful for “beta completion,” meaning quickly filling out your market exposure while you build out the single-name positions.
Electronic Trading Is Changing the Game
For a long time, bond trading was phone calls and instant messages. A trader would call a dealer, ask for a price, negotiate, and execute. Slowly, that is changing.
As of late 2020, about 38% of US investment-grade bond trading and 27% of high-yield trading happened on electronic platforms. That’s up significantly from five years earlier, though still far behind US interest rate markets at 65%.
The main platforms are Bloomberg, MarketAxess, TradeWeb, TruMid, and LiquidNet. Bloomberg dominates European corporate bonds. MarketAxess dominates North American corporate bonds.
Trading protocols on these platforms include anonymous requests for quotes (RFQs), disclosed RFQs, auctions, central limit order books, portfolio trades, and streaming prices. We’re still far from the stock market model of a fully electronic central limit order book. But progress is happening.
The New Issue Premium
When a company issues a new bond, something interesting happens. That bond doesn’t enter the major indices until the following month. During that gap, the bond tends to outperform.
Richardson shows that from 2016 to 2020, newly issued investment-grade bonds gained about 0.25% in their first couple of days. High-yield new issues gained about 0.85%. These returns were positive 72% and 86% of the time, respectively.
This is the “new issue concession.” Issuers slightly overpay (in yield terms) to attract buyers, and that extra yield translates into positive returns as the bond’s spread tightens after issuance.
For systematic investors, the primary market is a chance to buy bonds at the issue price without crossing any bid-ask spread. The catch is that you need scale and relationships with underwriting desks. You also need to prove you’re not a “flipper” who will dump the bond the next day.
Systematic Investors as Liquidity Providers
Here’s the part that really sets this chapter apart. Richardson argues that systematic investors have a unique advantage in bond markets, not just as buyers and sellers, but as liquidity providers.
Think about it. A systematic process has views on thousands of bonds at any given time. When a dealer shows an “axe” (an indication that they want to buy or sell a particular bond), a systematic investor can quickly evaluate whether that trade improves their portfolio. No committee meeting needed. No calling the PM for approval. The model already knows.
This matters because dealers increasingly operate as agents. They don’t want to hold inventory. They want to match buyers with sellers. A systematic investor who can respond quickly to axes, trade inside the bid-ask spread, and participate reliably in the primary market becomes a valuable counterparty.
The benefits go three ways. First, lower transaction costs because you trade inside the spread. Second, better alpha capture because you’re rebalancing continuously rather than at fixed intervals. Third, greater capacity because you’re using all available liquidity.
Transaction Cost Analysis
Richardson closes the chapter with a call for rigorous transaction cost analysis (TCA). The key insight is that looking only at the executed price versus the pre-trade mid-price understates your real costs.
Total shortfall includes the obvious trading cost (the spread you cross) plus slippage (how much the price moved between when your model generated the trade list and when you actually executed). Slippage can come from data processing delays, corporate action checks, or simply the time it takes for traders to work the order.
“Markout” analysis, which tracks how prices move after your trade, helps identify adverse selection. If prices consistently move against you after trading, something is wrong with either your information or your execution strategy.
The Bottom Line
Bond trading will never be as easy as stock trading. Too many bonds, too little standardization, too few natural counterparties. But systematic investors can turn these challenges into advantages. By having views on thousands of bonds, responding quickly to liquidity opportunities, and measuring every aspect of execution, they can actually become the liquidity that the market needs.
The market structure is evolving. Electronic platforms are growing. Dealer inventory is not coming back. The future of bond trading belongs to investors who can combine systematic models with smart execution.
Previous: Building a Fixed Income Portfolio
Next: Sustainability in Fixed Income Investing
Book: Systematic Fixed Income: An Investor’s Guide by Scott Richardson, Ph.D. Published by John Wiley & Sons, 2022. ISBN: 9781119900139.