Back in the summer of 2015, UBS released a note that highlighted an interesting trend.
Treasury market participants, the bank observed, were eschewing the cash market in favor of futures. Specifically, UBS looked at turnover in the cash market versus futs turnover. What they discovered wasn’t surprising, although it was slightly disconcerting if you care about market liquidity.
Here are some excerpts and a visual from the note:
For the past three months, daily average futures volume stands at nearly 70% of cash Treasuries, based on the notional amounts transacted. That is up from about 50% in 2011.The big leap in the turnover ratio occurred in 2014, and appears to have been sustained this year.
The trend was even more apparent when you drilled down. Have a look at the same analysis applied to short maturities:
So the takeaway there seems to be that from April of 2014 through at least June of 2015, market participants were attempting to dodge a lack of liquidity in cash markets by “migrating” (as UBS put it) to futures.
If you follow these types of things you might note that a similar dynamic is taking place in corporate bond markets where ETFs have become a substitute for the underlying assets.
These shifts could well create a self-feeding dynamic. That is, the more you rely on derivatives, the less you’re trading the cash markets. The less you trade the cash markets the more illiquid those markets become. The more illiquid the cash markets are, the more you turn to derivates as a more liquid alternative. And around we go. The “solution” (migrating to derivatives) makes the original problem (a lack of liquidity in the cash market) that much worse.”
So with that as the backdrop, consider the following piece from the NY Fed whose economists suggest that i) liquidity is drying up in cash markets for corporate bonds relative to CDS, and ii) the CDS market is becoming less liquid relative to CDS indices.
This would appear to be yet another incarnation of the dynamic described above. As the Fed writes, one explanation for the trend could be that “investment advisors specializing in corporate credit may have increased their long positions (by selling more protection) in CDX indexes in the second half of 2015 in anticipation of future fund outflows; by selling the more liquid CDS rather than buying the less liquid bonds, investment advisors still take on credit risk exposure while maintaining a relatively liquid portfolio.”
“The strategic substitution away from bonds to CDS further decreases liquidity in the bond market relative to the CDS market, contributing to the widening CDS-bond basis,” the Fed adds, echoing what I said above about the apparent circularity problem.
Of course another explanation is a less forgiving regulatory regime. “Market participants have also suggested that post-crisis regulatory changes have limited the willingness of regulated institutions to engage in arbitrage trades across a variety of markets,” the Fed goes on to note.
While what you read below may seem esoteric, it’s important that you at least grasp the premise.
Corporate bonds are an important source of funding for public corporations in the United States. When these bonds cannot be easily traded in secondary markets or when investors cannot easily hedge their bond positions in derivatives markets, the issuance costs to corporations increase, leading to higher overall funding costs. In this post, we examine recent trends in arbitrage-based measures of liquidity in corporate bond and credit default swap (CDS) markets and evaluate potential explanations for the deterioration in these measures that occurred between the middle of 2015 and early 2016.
Two Credit Arbitrage Trades
In the first trade we consider, the CDS-cash bond “arbitrage trade,” an investor buys a corporate bond and simultaneously buys CDS protection on the bond (or vice versa on both legs of the trade). The CDS-bond basis is the difference between the CDS spread and the implied spread on the bond. “Arbitrage” refers to the fact that the spread of the bond should equal the CDS spread for the same firm if they represent only the firm’s default risk. In practice, the basis is generally not zero. One interpretation of this basis is that it measures the liquidity of the cash bond market relative to the CDS market; when one market is more liquid than the other, investors will accept a lower spread as compensation for the greater liquidity.
The chart below plots the CDS-bond basis for investment-grade (left panel) and high-yield (right panel) bond indexes since January 2005. The CDS-bond basis has been increasing since January 2015 for investment-grade bonds and since the middle of 2015 for high-yield bonds, suggesting that liquidity of the cash bond market has been deteriorating relative to the CDS market.
In the second trade we consider, the CDX-CDS arbitrage trade, an investor buys protection on a CDX index and sells protection on the portfolio of single-name CDS contracts that replicates the index (or vice versa). The resulting CDX-CDS basis equals the absolute value of the difference between the CDX index spread and the weighted average spread paid on the replicating portfolio of single-name CDS. Recent academic research has argued that the CDX-CDS basis measures the overall liquidity of the CDS market, with changes to the basis accounting for 30 percent of CDS returns on average. The next chart plots the quoted spread, the single-name implied spread, and the CDX-CDS basis for the North American investment-grade (left panel) and North American high-yield (right panel) on-the-run CDX indexes. Similar to the CDS-bond basis, the CDX-CDS basis has been increasing since the beginning of 2015, suggesting that while liquidity of the CDS market has improved relative to the cash bond market, it has deteriorated relative to the CDX market. In February 2016, the CDX-CDS basis reached peaks not seen since the financial crisis, reaching a third of the financial crisis peak.
The moves in the credit bases between the second half of 2016 and the first quarter of 2016 were abnormal relative to this historical experience. The table below shows that the one-month change between January and February 2016 in the CDS-bond basis was in the bottom (most negative) fifth percentile of one-month changes observed between January 2005 and April 2016, and the six-month change (between August 2015 and February 2016) was in the bottom tenth to fifteenth percentile of historical changes for the investment-grade and high-yield indexes. For the CDX-CDS basis, both the one-month and the six-month changes are in the highest (largest) fifth percentile of the historical distribution of changes for both the investment-grade and the high-yield indexes.