FINRA found that liquidity in the U.S. corporate-bond market is healthy “by most measures.” FINRA’s Research Note analyzed all Trade Reporting and Compliance Engine (“TRACE”) transactions from 2003 to September 2015.
FINRA’s research found that:
Most measures indicate a healthy market: new bond issuance is at a “record level,” transaction volumes have continued to grow and the number of trades is rising. The cost of trading corporate bonds has been decreasing, as indicated by narrower bid-ask spreads and falling price impact to block trades.
However, several measures offer evidence of potentially significant changes in how the market is working, including smaller average trade size and a declining proportion of bonds traded in blocks of $5 million or more. “These trends are consistent with a market that has a larger number of issues, more electronic trading and a growing network of counterparties.”
Lofchie Comment: The numbers presented in the FINRA report send considerably more mixed or ambiguous signals than the tone of the report would suggest. The report certainly has some positive numbers in it, but a closer look at those numbers, when viewed in context, may present a different picture. Take, for example, the finding that bid/ask spreads had declined, which seems to be a healthy indicator for the market. The estimated bid/ask spreads are analyzed in the study without regard to average trade size. But as the average trade size has declined substantially, one would expect the bid/ask spread to decline as well. It is not obvious that the bid/ask spread has declined more than one would expect from the correlating decline in the trade size.
The main not so positive number reported is to the effect that turnover in bonds is slower, with the result that it takes longer to trade out of a position. On its face, this negative fact seems inconsistent with the decline in the bid/ask spread. If the bid/ask spread is down, shouldn’t it be easier/quicker to get out of positions? Likewise, the decline in volume seems inconsistent with the fact that short-term electronic traders now play a much larger role in the bond markets. Given that one could expect these traders to accumulate higher volume, as they do not hold positions for an extended period, one would think that trading volume should have risen materially, not declined. In summary, a decline in bid/ask spreads could not be considered such good news to the extent that it is combined with a decline in trade size; and a decline in trade volume is somewhat more negative news if that volume was increased by short-term electronic traders who may be expected to pump volume up.
The report also declared that it was a positive sign for the workings of the bond markets that the amount of new issuance was high. However, as the study points out, the cost of borrowing is relatively close to zero, so one would expect new issuances to be high. It is not obvious why that should be viewed as a positive sign for how well the bond trading markets function.
None of the above is meant to say that the numbers in the report tell an inherently or completely negative story; only that they and their relationships seem to send considerably more mixed or ambiguous signals than the report suggests.
Update: The Wall Street Journal of Dec. 14, 2015, has a story titled “Junk Bonds Stagger as Funds Flee.” The story reports 10% spreads between bids and offers, a stark contrast to the findings and tone of the report.