PrefBlog

Here’s an interesting paper partly about TIPS liqudity, titled The Microstructure of the TIPS Market by Michael J. Fleming and Neel Krishnan:

• • The potential advantages of Treasury inflationprotected securities have yet to be fully realized, mainly because TIPS are not as liquid as nominal Treasury securities.
• • The less liquid nature of TIPS may adversely affect prices relative to those of nominal securities, offsetting the benefits of TIPS having no inflation risk.
• • A study of TIPS, using novel tick data from the interdealer market, provides new evidence on the liquidity of the securities and how liquidity differs from that of nominal securities.
• • Analysis of various liquidity measures suggests that trading activity and the incidence of posted quotes may be better cross-sectional gauges of TIPS liquidity than bid-ask spreads or quoted depth.
• • Differences in intraday trading patterns and announcement effects between TIPS and nominal securities likely reflect the different use, ownership, and cash-flow attributes of the securities

…
These potential benefits have not been fully realized, mainly because TIPS lack market liquidity compared with nominal securities.{2} This lack of liquidity is thought to result in TIPS yields having a liquidity premium relative to nominal securities, which offsets the inflation risk premium.{3} Similarly, the presence of a liquidity premium in TIPS yields complicates inferences of inflation expectations, particularly if the premium changes over time. However, despite the importance of TIPS liquidity and the market’s large size ($728 billion as of November 30, 2011), there has been virtually no quantitative evidence on the securities’ liquidity.

Footnote 2: Market liquidity is defined here as the cost of executing a trade, which can depend on the trade’s size, timing, venue, and counterparties. It is often gauged by various measures, including the bid-ask spread, the price impact of trades, quoted depth, and trading activity.

Footnote 3: D’Amico, Kim, and Wei (2008) estimate that the liquidity premium was about 1 percent in the early years of the TIPS program. Pflueger and Viceira (2011) find that the liquidity premium is around 40 to 70 basis points during normal times, but was more during the early years of TIPS and during the 2008-09 financial crisis. Sack and Elsasser (2004) argue that TIPS have not reduced the Treasury’s financing costs because of several factors, including lower liquidity. Roush (2008) finds that TIPS have saved the government money, except during the early years of the program. Dudley, Roush, and Ezer (2009) show that the ex ante costs of TIPS issuance are about equal to the costs of nominal securities issuance.
…
Our study proceeds as follows. Section 2 discusses institutional features of the market for TIPS. In Section 3, we describe the tick data used in our empirical analysis. Section 4 reports our empirical results, including trading activity by sector, the liquidity of on-the-run and off-the-run securities, price impact estimates, intraday patterns in trading activity and liquidity, and the effects of major announcements. Section 5 concludes.
…
Our analysis of the TIPS market identifies several microstructure features also present in the nominal Treasury securities market, but several unique features as well. As in the nominal market, there is a marked difference in trading activity between on-the-run and off-the-run TIPS, as trading drops sharply when securities go off the run. In contrast to the nominal market, there is little difference in bid-ask spreads or quoted depth between these securities, but there is a difference in the incidence of posted quotes. The results suggest that trading activity and quote incidence may be better crosssectional measures of liquidity in the TIPS market than bid-ask spreads or quoted depth.

Intraday patterns of trading activity are broadly similar in the TIPS and nominal markets, but TIPS activity peaks somewhat later, likely indicating differences in the use and ownership of these securities. Announcement effects are also different, probably reflecting the types of information most important to the particular securities. The employment report is the most important announcement in the nominal market, but it elicits relatively little response in the TIPS market in terms of trading activity. In contrast, announcements of the consumer price index and the results of TIPS auctions precipitate significant increases in TIPS trading activity, likely indicating these announcements’ particular importance to TIPS valuation

There’s also Trading Activity and Price Transparency in the Inflation Swap Market by Michael J. Fleming and John R. Sporn:

• • Liquidity and price transparency in derivatives markets have become increasingly important concerns, yet a lack of transaction data has made it hard to fully understand how the inflation swap and other derivatives markets work.
• • This study uses novel transaction data to shed light on trading activity and price transparency in the rapidly growing U.S. inflation swap market.
• • It reveals that the market is reasonably liquid and transparent, despite its over-the-counter nature and low level of trading activity. Transaction prices are typically near widely available end-of-day quoted prices and realized bid-ask spreads are modest.
• • The authors also identify concentrations of activity in certain tenors and trade sizes and among certain market participants as well as point to various attributes that explain trade sizes and price deviations.

…
Several recent studies have compared the inflation swap rate with breakeven inflation as calculated from Treasury inflationprotected securities (TIPS) and nominal Treasury bonds.1 The two market-based measures of expected inflation should be equal in the absence of market frictions. In practice, inflation swap rates are almost always higher, with the spread exceeding 100 basis points during the recent financial crisis.
…
Our data set contains 144 U.S. dollar zero-coupon inflation swap transactions, or an average of 2.2 transactions over the 65 trading days in our sample.9 Daily notional trading volume is estimated to average $65 million. Three-quarters (108/144) of the transactions are new trades, 24 percent (35/144) are assignments of existing transactions (whereby one counterparty to a swap steps out of the deal and assigns its position to a new counterparty), and 1 percent (1/144) are cancelations. One new transaction has a forward start date, for which the accrual period begins two years after the trade date, with the remaining 107 new transactions starting two or three business days after the trade date.
…
We also identify a concentration of activity among certain market participants. In particular, 54 percent (78/144) of our trades are between G14 dealers, 39 percent (56/144) are between G14 dealers and customers, and 7 percent (10/144) are between customers. Of the new trades between G14 dealers and customers, the G14 dealer receives fixed 63 percent (19/30) of the time and pays fixed 37 percent (11/30) of the time.11 New trades in which dealers receive fixed are larger, so that dealers receive fixed for 81 percent of new contract volume. That is, dealers are largely paying inflation and receiving fixed in their interactions with customers.
…
Our analysis of a novel transaction data set uncovers relatively few trades—just over two per day –in the U.S. zero-coupon inflation swap market. Trade sizes, however, are large, averaging almost $30 million. Sizes are generally larger for new trades, especially if they are bulk and allocated across subaccounts, and tend to decrease with contract tenor. We also identify concentrations of activity—with 45 percent of trades at the ten-year tenor, and 36 percent of all trades (and 48 percent of new ones) for a notional amount of $25 million. Over half the trades (54 percent) are between G14 dealers, 39 percent are between G14 dealers and other market participants, and 7 percent are between other market participants. We identify just eighteen market participants during our study’s sample period, made up of nine G14 dealers and nine other market participants.

Despite the low level of activity in this over-the-counter market, we find that transaction prices are quite close to widely available end-of-day quoted prices. The differential between transaction prices and end-of-day quoted prices tends to decrease with tenor and increase with trade size and for customer trades. By comparing trades for which customers pay fixed with trades for which they receive fixed, we are able to infer a realized bid-ask spread for customers of 3 basis points, which is consistent with the quoted bid-ask spreads reported by dealers.

In sum, the U.S. inflation swap market appears reasonably liquid and transparent despite the market’s over-the-counter nature and modest activity. This likely reflects the fact that the market is part of a larger market for transferring inflation risk that includes TIPS and nominal Treasury securities. As a result, inflation swap positions can be hedged quickly and with low transaction costs using other instruments, and prices of these other instruments can be used to efficiently price inflation swaps, despite modest swap activity

Not exactly the world’s biggest market! I looked up inflation swaps because I was interested in the question “Who the hell pays inflation”, which came to mind due to this article in the Globe, The government ditched inflation-protected bonds – companies should start issuing their own by JOHN H. COCHRANE AND JON HARTLEY:

If the government won’t do it, corporations, banks and financial institutions should issue these bonds themselves rather than just complain. Not every asset must be provided by the government.
…
If the government won’t do it, however, there is no reason that the government’s critics can’t issue them. Companies can issue real return bonds, as they already issue U.S. dollar bonds. Banks can offer real return accounts and certificates of deposit.

If the government steps out of the market, there’s all the more demand for private issuers to step in. Pension funds desperate to replace vanishing inflation-indexed government bonds are natural clients. Company profits rise and fall with inflation, so they have a natural incentive to issue bonds whose payments rise and fall with inflation. Even mortgage rates could rise and fall with an index of wages.

Why not? Broadly, this reluctance seems one more symptom of an overleveraged, overregulated, government-dependent and not very competitive or innovative banking and financial system. Banks and other financial institutions only want to issue or expand a new product if they can quickly lay off the risk onto the government, and earn steady fees. The model of issuing equity to bear risk and then offering a profitable innovative product to consumers is too out of fashion.

Frankly, I thought the article was naive, but thought: “Who the hell would issue these things? Who’s got a natural hedge against inflation that they might want to offload? Assuming they can recover the ultra-massive liquidity premium there’s gonna be on a, say, 1-billion long-term linker issue from a corporation, that is.” All I could think of was utility companies who have long-term assets currently financed by long-term nominal bonds, with the assets producing commodity-linked revenue. Maybe they could finance with linkers instead? Maybe pipelines? So, I started looking for information on inflation swaps …

I can’t answer the question definitively. The authors of the swaps paper didn’t investigate where the open interest is lodged. But there is enough information in the paper that I’m willing to bet a nickel (a full nickel, mind you, not just a few pennies) that it’s the dealers. The dealers will pay inflation and they buy TIPS to hedge. BORRRRRRR-ING! And it doesn’t work without government-issued linkers.

The Bank of Canada has released Staff Discussion Paper 2023-2, by Monica Jain, Walter Muiruri, Jonathan Witmer, Sharon Kozicki & Jeremy Harrison titled Summaries of Central Bank Policy Deliberations: A Canadian Context:

This paper provides the context, rationale and key considerations that informed the Bank of Canada’s decision to publish a summary of monetary policy deliberations. It includes an analysis of how other central banks disclose minutes and summaries of their monetary policy deliberations.

Most other central banks surveyed publish some sort of summary of deliberations. The Bank of Canada’s existing communications already include aspects of these summaries. However, the Bank does not normally provide some information that they contain, such as:

• • a review of the policy choices that were discussed
• • a diversity of viewpoints on the economic outlook and policy choices
• • the perspectives of individual members

Publishing a summary of deliberations could enhance transparency, accountability and credibility and also reinforce the Bank’s independence. However, these benefits must be balanced against the potential for constraints on internal debate or the sending of mixed messages about the Bank’s outlook and decisions. The Bank of Canada Act empowers the Governor to make decisions, but in practice, decisions are made by consensus among members of the Bank’s Governing Council. This decision-making by consensus could have implications for what could or should be included in a summary.

In the Canadian context, assuming the Bank will provide additional information, we also discuss some advantages and disadvantages of providing a summary of deliberations as a separate communication product or as an enhancement to current communications products.

The material in the paper originally served as background information for internal discussions at the Bank of Canada around publishing a summary of policy deliberations. Following those discussions, the International Monetary Fund (IMF) published a review of the Bank of Canada’s transparency, concluding that the Bank “… sets a high benchmark for transparency” (IMF 2022). In that review, the IMF provided a recommendation on how the Bank could further improve its transparency by providing more information on its monetary policy deliberations. In response to the IMF review and internal discussions at the Bank, the Bank has publicly committed to providing a summary of its policy deliberations beginning in February 2023.

The most desperately needed disclosure is – as Assiduous Readers will be sick to death of me complaining – voting records. So here’s a table comprised of their summaries of voting records:

Consensus is for second-raters and time-servers. A confident, intelligent person will not feel any shame about being in the minority, even if on a repeated basis. Hell, Leon Trotsky was a proud member of the Menshevik (minority) Party and he got a lot of respect in his day! I take issue with the following quotation from the abstract:

However, these benefits must be balanced against the potential for constraints on internal debate or the sending of mixed messages about the Bank’s outlook and decisions.

Dammit, I want mixed messages! Only idiots will take the view that monetary policy is a puzzle with only one answer – it’s complex and is concerned exclusively of forecasts about the future that are, we hope, backed up by excellent data and analysis of current conditions. While the consensus phrase ‘risks to the forecast include…’ may attempt to give a sense of the uncertainty, it is nowhere near as useful as ‘so-and-so was so concerned about the potential for X that he voted against the policy decision! He put his name on it! He stepped up and advocated an unpopular position for no other reason than he thought it was right! Pay attention, people!’

I will also take issue with the other justification put forward, that increased transparency (such as publicizing voting records) will constrain internal debate. OK, I say, relative to what? People will feel constrained from vigorously asserting their views for all sorts of stupid reasons and I will suggest that the necessity for eventual consensus is a greater constraint that the publication of a dissenting vote with a brief note of explanation. Arse-kissers and group-thinkers thrive in an environment in which they are explicitly expected to agree with the loudest voice in the meeting, and we don’t want any of them setting monetary policy!

Other data compared in the tables are disclosures of:

• Discussion of risks
• Data and projections
• Financial developments
• Economic developments
• Areas of discussion in deliberations specified
• Detail of meeting transcript/summary
• Diversity of views
• Indications of future policy interest rate decisions
• Indications of future non-interest-rate policy decisions
• Publishes a monetary policy report
• Discusses conflicts in policy decisions

Sarah Breeden, the Bank of England’s Executive Director for Financial Stability Strategy and Risk, has delivered a speech titled Risks from leverage: how did a small corner of the pensions industry threaten financial stability?:

But in the days leading up to that fateful Wednesday and following the announcement of the Government’s growth plan on 23 September, long-dated gilt yields in particular had moved with extraordinary and unprecedented scale and speed.

Now volatility itself does not warrant Bank of England intervention. Indeed, it’s essential that market prices are allowed to adjust to changes in their fundamental determinants efficiently and without distortion.

However, some liability-driven investment (LDI) funds were creating an amplification mechanism in the long-end of the gilt market through which price falls had the potential to trigger forced selling and thereby become self-reinforcing. Such a self-reinforcing price spiral would have resulted in even more severely disrupted gilt market functioning. And that would in turn have led to an excessive and sudden tightening of financing conditions for households and businesses.

In response to this threat, the Bank of England intervened on financial stability grounds. But what led to that intervention?

The root cause is simple – and indeed is one we have seen in other contexts too – poorly managed leverage.
…
Many UK DB pension schemes have been in deficit, meaning their liabilities – their commitments to pay out to pensioners in the future – exceed the assets they hold. DB pension schemes invest in long-term bonds to hedge the interest rate and inflation risk that arises from these long-term liabilities. But that doesn’t help them to close their deficit. To do that, they invest in ‘growth assets’, such as equities, to get extra return to grow the value of their assets. An LDI strategy delivers this, using leveraged gilt funds to allow schemes both to maintain material hedges and to invest in growth assets. Of course that leverage needs to be well managed.

The rise in yields in late September – 130 basis points in the 30-year nominal yield in just a few days – caused a significant fall in the net asset value of these leveraged LDI funds, meaning their leverage increased significantly. And that created a need urgently to delever to prevent insolvency and to meet increasing margin calls.

The funds held liquidity buffers for this purpose. But as those liquidity buffers were exhausted, the funds needed either to sell gilts into an illiquid market or to ask their DB pension scheme investors to provide additional cash to rebalance the fund. Since persistently higher interest rates would in fact boost the funding position of DB pension schemes[1], they generally had the incentive to provide funds. But their resources could take time to mobilise.

The issue was particularly acute for one small corner of the LDI industry – pooled funds. In these funds, which make up around 10-15% of the LDI market, a pot of assets is managed for a large number of pension fund clients who have limited liability in the face of losses. The speed and scale of the moves in yields far outpaced the ability of the large number of pooled funds’ smaller investors to provide new funds who were typically given a week, in some cases
two, to rebalance their positions. Limited liability also meant that these pooled fund investors might choose not to provide support. And so pooled LDI funds became forced sellers of gilts at a rate that would not have been absorbed in normal gilt trading conditions, never mind in the conditions that prevailed during the stressed period.

Other LDI funds, with segregated mandates, were more easily able to raise funds from their individual pension scheme clients. However, given their scale, at 85-90% of the market, some of these funds were also contributing to selling pressure, making the task at hand for pooled LDI funds even harder. And of course if the pooled funds had defaulted, the large quantity of gilts held as collateral by those that had lent to the funds would potentially be sold on the market too.

With the gilt market unable to absorb such forced sales, yields would have been pushed even higher, making the scale of the selling need even larger still. This is the self-reinforcing spiral that the Bank intervened to prevent.

The Bank’s 13 day and £19.3 billion intervention was made on financial stability grounds. It was the first example of us acting to deliver our financial stability objective through a temporary, targeted intervention in the gilt market.

But let me emphasise: the asset purchases were a means to an end. They were designed to create the right conditions in the right part of the gilt market for long enough so that the LDI funds could build resilience so that their leverage would be well managed once the asset purchases had ceased and should gilt market instability return.
…
A common factor across all the uses of leverage I have just described is that it can increase the exposure of the leverage taker to underlying risk factors – whether that be house prices, earnings, interest rates, currencies or asset prices. It follows therefore that leverage can amplify shocks to each of these risk factors. And in a stress, that can lead both to sudden spikes in demand for liquidity – either to support the financing of leveraged positions or as deleveraging leads to forced sales – and a corresponding contraction in liquidity supply, with potentially systemic consequences.

Leverage is of course not the only cause of systemic vulnerability in the non-bank system – as we have seen with liquidity mismatch driving run dynamics in money market funds (MMFs) and open-ended funds (OEFs) during the dash for cash.[4] But it is important where any form of leverage is core to a non-bank’s business and trading strategy. Indeed what happened to LDI funds is just the latest example of poorly managed non-bank leverage throwing a large rock into the pool of financial stability. From Long Term Capital Management in 1998; to the 2007 run on the repo market; to hedge fund behaviour in the 2020 dash for cash; and the failure of Archegos in 2021.

These episodes highlight the need to take into account the potential amplifying effect of poorly managed leverage, and to pay attention to non-banks’ behaviours which, particularly when aggregated, could lead to the emergence of systemic risk.
…
Regulators worked with LDI funds during the Bank’s operations to ensure greater resilience for future stresses. And in aggregate, intelligence suggests that LDI funds raised over £40 billion in funds and made over £30 billion of gilt sales during our operations, both of which have contributed to significantly lower leverage.

As a result, LDI funds report that their liquidity buffers can withstand very much larger increases in yields than before, well in excess of the previously unprecedented move in gilt yields. And so the risk of LDI fund behaviour triggering ‘fire sale’ dynamics in the gilt market and self-reinforcing falls in gilt prices is – for now at least – significantly reduced. It is important that it stays that way.

I’m sure there will be more material on this liquidity black hole to follow, but for now I’ll just register my continuing disgust with the charlatans and nincompoops who are such a feature of the investment management industry.

A discussion on an unrelated thread regarding historical pricing on brokerage statements for GICs eventually expanded to include historical pricing for Money Market Funds. As MMFs are marketable instruments, there are wider implications of this policy than there are for GICs.

Jonathan Witmer of the BoC wrote a working paper in 2012 titled Does the Buck Stop Here? A Comparison of Withdrawals from Money Market Mutual Funds with Floating and Constant Share Prices:

Recent reform proposals call for an elimination of the constant net asset value (NAV) or “buck” in money market mutual funds to reduce the occurrence of runs. Outside the United States, there are several countries that have money market mutual funds with and without constant NAVs. Using daily data on individual fund flows from these countries, this paper evaluates whether the reliance on a constant NAV is associated with a higher frequency of sustained fund outflows. Preliminary evidence suggests that funds with a constant NAV are more likely to experience sustained outflows, even after controlling for country fixed effects and other factors. Moreover, these sustained outflows in constant NAV money market funds were more acute during the period of the run on the Reserve Primary fund, and were subdued after the U.S. Treasury guarantee program for money market funds was put in place. Consistent with the theory that constant NAV funds receive additional implicit support from fund sponsors, fund liquidations are less prevalent in funds with a constant NAV following periods of larger outflows.
…
This paper is the first to examine the usage of a constant NAV structure across countries. It is well known that money market funds in some countries, such as the United States, employ a constant NAV structure. It is less well known to what extent other countries use a different structure. The main difference between floating NAV and constant NAV money market funds is the use of amortized cost accounting. Floating NAV money market mutual funds measure the value of their positions using fair value or market prices. For constant NAV money market funds, the value is recorded as the initial cost, plus the straight line amortization of the position’s premium or discount at the time of purchase through to the position’s maturity date. This paper shows that many European countries have a mixture of both fund types.

Here’s the interesting bit – how predatory traders are able to fleece naive investors:

This paper also contributes to the broader literature that examines the relation between stale share prices, illiquid fund holdings, and fund flows in equity and bond mutual funds. Arbitrageurs can take advantage of stale prices in illiquid mutual funds at the expense of the remaining shareholders. These apparent arbitrage opportunities induce a change in flows in these mutual funds. The paper by Lyon (1984) finds this arbitrage activity dilutes other shareholders in money market funds by an estimated 10 bps per year. This dilution is even larger in international equity mutual funds, where dilution can be upwards of 1% per year (e.g., Greene and Hodges, 2002; Zitzewitz, 2003).
…
During the first part of September 2008 when there was a run on the Reserve Primary Fund, constant NAV money market funds experienced more outflows than did floating NAV money market funds. Further, after the U.S. Treasury implemented its guarantee program for money market funds, constant NAV U.S.-domiciled U.S. dollar funds performed much better and sustained a decrease in prolonged outflows during the guarantee period, relative to non-U.S. domiciled U.S. dollar funds.
…
After the crisis, the SEC amended rule 2a-7 to improve the resiliency of money market mutual funds. These amendments included tighter restrictions on the credit quality, maturity, and liquidity of portfolio holdings for money market funds. The maximum dollar-weighted average maturity was reduced to 60 days, and a maximum dollar-weighted average life to maturity was introduced and set at 120 days. As for the liquidity requirements, a minimum of ten percent of a fund’s portfolios must be invested in “Daily Liquid Assets” and a minimum of thirty percent must be invested in “Weekly Liquid Assets”. The amended rule 2a-7 also requires monthly website disclosure of portfolio holdings, including information

The author concludes, in part:

This paper has several important policy implications. There is an active push to reform money market mutual funds in the wake of the financial crisis and more specifically following the run on the Reserve Primary Fund and subsequent government support of money market funds in the United States. One of the primary proposals is to move away from the CNAV money market fund structure and towards the VNAV structure. Some observers have contended that such a move does little to reduce the occurrence of runs in money market mutual funds, based on anecdotal evidence of run behaviour in ultrashort bond funds in the United States and enhanced money market funds in Europe, both of which maintain a VNAV structure (Investment Company Institute, 2011; HSBC, 2011). These funds, however, are not subject to the same liquidity, credit, and maturity restrictions as money market funds. This paper compares a large number of money market mutual funds across several countries and finds that, on the contrary, the VNAV structure is less susceptible to run-like behaviour relative to CNAV money market funds.

However, the VNAV structure does not fully eliminate this run-like behaviour. This is consistent with the model of Chen, Goldstein, and Jiang (2011), which shows that mutual funds holding illiquid assets experience more outflows following a period of poor performance, relative to funds holding liquid assets (their empirical examination focuses on equity mutual funds). That is, in their model investors may redeem on the self-fulfilling belief that others will be redeeming, imposing the costs of liquidating the fund’s illiquid assets on remaining shareholders. While money market funds generally hold liquid, shortterm assets, these assets may become illiquid during periods of stress or, put another way, during periods when there is a belief that a fire sale of some money market fund holdings may occur. Even during periods of stress, however, CNAV money market funds are more prone to run-like behaviours, relative to VNAV money market funds.

Given my own views on the subject, expressed in A Collateral Proposal and The Future of Money Market Regulation, I was most interested in his final paragraph:

Not only does the CNAV structure have a higher occurrence of sustained outflows, but also there is some evidence to suggest that it is associated with an implicit guarantee provided by fund sponsors. This implicit guarantee has both advantages and disadvantages. The presence of an implicit guarantee can reduce moral hazard and reduce risk-taking in money market mutual funds, since the fund sponsor would be concerned that the poor performance of the fund may have negative spillovers on the sponsor’s other businesses (Kazpercyk and Schnabl, 2012). The amount of risk-taking depends upon both the sponsor’s financial strength as well as the reputational concerns about the effect of “breaking the buck” on the rest of the sponsor’s fund and non-fund businesses. On the other hand, an implicit guarantee is a potential channel for contagion between the banking sector and money market mutual funds. Losses in a money market mutual fund may be passed onto the fund sponsors should they provide support to the fund. As well, a weakening of a fund sponsor could be passed onto the money market fund sector through a reduction in the value of the implicit guarantee.

Forward interest rates have emerged as a bone of contention in the analysis of the proposed TransAlta preferred share exchange offer, so as part of the preparation for my promised weekend post, I’ll post a few links to some papers that illustrate why the Expectations Hypothesis cannot be used as a predictor.

Joseph R. Dziwura and Eric M. Green wrote a paper in 1996 for the New York Fed titled Interest Rate Expectations and the Shape of the Yield Curve:

According to the rational expectations hypothesis of the term structure (REHTS) long term rates should reflect market expectations for the average level of future short-term rates. The purpose of this paper is to examine whether REHTS assumptions conform to the term structure of outstanding U. S. Treasury securities from 1973 to 1995, and to examine the behavior of term premiums and to what extent they influence the shape of the forward curve. REHTS assumptions are re-examined using familiar regression tests to determine the forecast power of forward rates for subsequent spot rates, and we use excess holding period returns, the extra return earned on a security sold prior to maturity, as the ex poste measurement of the term premium. We find that forward rates explain only some of the variance in future spot rates, the forecast power of forward rates varies with maturity, and the term premia is time-varying. We decompose the forward rate into the current spot rate, a term premium, and an expected interest rate change, where the term premium is the sum of a risk premium and a convexity premium. We find that on average term premiums have contributed more to the shape of the forward curve than have expected rate changes, and find that expected and past interest rate volatility, as well as the slope of the yield curve, may provide information on the size of expected term premiums.

Another paper was by Massimo Guidolin and Daniel L. Thornton of the St. Louis Fed, titled Predictions of Short-Term Rates and the Expectations
Hypothesis:

Despite its role in monetary policy and finance, the expectations hypothesis (EH) of the term structure of interest rates has received virtually no empirical support. The empirical failure of the EH has been attributed to a variety of econometric biases associated with the single-equation models most often used to test it; however, none of these explanations appears to account for the massives [sic] failure reported in the literature. We note that traditional tests of the EH are based on two assumptions—the EH per se and an assumption about the expectations generating process (EGP) for the short-term rate. Arguing that convential [sic] tests of the EH could reject it because the EGP embedded in these tests is significantly at odds with the true EGP, we investigate this possibility by analyzing the out-of-sample predictive prefromance [sic] of several models for predicting interest rates and a model that assumes the EH holds. Using standard methods that take into account parameter uncertainty, the null hypothesis of equal predictive accuracy of each models relative to the random walk alternative is never rejected.

One may hope their work is more reliable than their proof-reading!

Intuitive Analytics is a financial software firm which has published a blog-post by Peter Orr titled 50 Years of UST Yields – How Well do Forwards Predict? that was exactly what I was looking for:

As we’ve written on these pages before, forecasting is a necessary evil in finance. It’s uncertain by nature and of course the longer the horizon, the more difficult the job. The theory that forward rates are good predictors of future realized rates is called the expectations hypothesis and as one MIT professor put it, “If the attractiveness of an economic hypothesis is measured by the number of papers which statistically reject it, the expectations theory of the term structure is a knockout.”

For fun (and to dust off my fast fading coding skills) I went back and looked at how US Treasury implied forward 10Y rates have done in forecasting realized 10Y UST yields from July, 1959 to the present. We used first of month data for 3, 6 and 12 month Tbills as zero rates (making the appropriate daycount adjustments of course) and then 2, 3, 5, 7, 10, 20, and 30-year UST coupon instruments for our implied 10Y forward calculations. And this is what we get…

Click for Big

The red line is the actual 10Y yield over the period and the “hair” is the implied 10Y par yield 1, 2, 3, and 5 years forward. The way to read this then is to look at how often the hair tracks with the actual realization of the 10Y yields as shown by the red line. In general, during this single big rate cycle we’ve seen over the last 50 years, forward rates have badly underpredicted when rates were going up (note the implied decreasing 10Y forwards during the 70s) and then overpredicted over the last 30 or so years as rates have fallen. How badly do forwards do? Well over this 50 year span, and this holds over most subperiods as well, you’d be better off as a forecaster just assuming today’s yield curve stays constant i.e. a perfectly random walk.

A Bloomberg piece titled How to Lose $667 Million in Bond Trades Without Trying discusses why stupid and lazy portfolio managers underperform:

Bond investors can waste a lot of money and not even know it.

They lost about $667 million in the year ended March 31 by paying higher prices for corporate bonds that were available at lower prices elsewhere, according to September research by Larry Harris, a business professor at the University of Southern California.

In most of the deals the investors simply did not know that the lower prices existed because they rely on human traders to tell them the value of bonds at any given moment before they make a trade. (Not to mention the salaries they need to pay those brokers to work the phones to find out who holds what and who might want to sell.)

The author, Lisa Abramowicz, mentions regulatory efforts to destroy the corporate bond market:

So far, the Securities and Exchange Commission is only encouraging big bond firms to use electronic marketplaces more frequently so that investors have an easier way to see market prices in real time.

But if that doesn’t work, regulators may take more invasive measures to streamline the playing field and make it cheaper to do business in the $8 trillion market for U.S. company bonds.

Regardless of what the SEC might do, it makes sense for Wall Street banks to work together to find a more efficient way to trade bonds because it may be the best for their bottom lines.

All this is interesting in light of the pending crippling of the Canadian corporate bond market discussed last week. Ms. Abramowicz buttresses her views – and regulators are virtually certain to follow her – by referencing a recent paper by Lawrence Harris of the University of Southern California titled Transaction Costs, Trade Throughs, and Riskless Principal Trading in Corporate Bond Markets:

This study analyzes the costs of trading bonds using previously unexamined quotations data consolidated across several electronic bond trading venues. Much bond market trading is now electronic, but the benefits largely accrue to dealers because their customers often do not trade at the best available prices. The trade through rate is 43%; the riskless principal trade (RPT) rate is above 42%; and 41% of customer trade throughs appear to be RPTs. Average customer transaction costs are 85 bp for retail-size trades and 52 bp for larger trades. Estimated total transaction costs for the year ended March 2015 are above $26 billion, of which about $0.5 billion is due to trade-through value while markups on customer RPTs transfer $0.7M to dealers. Small changes in bond market structure could substantially improve bond market quality.

The problem, as is usual with this type of paper, lies in the assumption of the very first sentence of the introduction:

Brokers are supposed to obtain the best available prices for their clients.

In virtually all cases in the bond market, the dealer is acting as principal. It is not just his privilege, but his job to leave his counterparties naked, hungry and freezing. This fundamental misstatement of the facts of the transaction persists throughout the paper. Particularly disgusting is the claim:

Although this transaction might not strictly be a trade through (it would not be if the broker-dealer exhausts all the size at the quoted price), the broker-dealer clearly is front-running the customer order, though not necessarily illegally.

Front-running is a breach of trust and can occur only when the intermediary is an agent of the trade initiator, therefore having a fiduciary responsibility to the initiator. The concept does not apply to trades executed as principal.

Another problem is with his definition of “transaction costs”:

I estimate the cost of trading for the side that initiated the trade by first identifying that side, and then by comparing the trade price to the quote midpoint price.

This definition makes dealer markups appear worse than they actually are.

Markups and commissions both contribute to transaction costs. Markups are incorporated in the price whereas commissions are tacked onto the price. Both allow brokers to recover the costs of arranging trades, and presumably all other costs of providing trading services to their clients.
…
Markups differ from commissions because broker-dealers generally do not fully disclose markups to their clients.

They also differ from commissions in that commissions apply to agency trades while markups apply to principal trading.

Even when broker-dealers fully disclose the nature of their relationships with their clients—that they are acting as principle [sic] and not as agent—many clients may not recognize the distinction and its implications. The distinction can be difficult to recognize when the broker-dealer sometimes acts as broker and sometimes as dealer, a process commonly called dual trading.

If clients do not recognize the distinction then they should not be trading. Traders in the institutional market will almost always be professionals and will have passed numerous proficiency tests set by the regulators. If retail traders want to play with the big boys and trade individual bonds themselves, they should recognize that step one is learning the rules of the game.

I will admit to long-term confusion over this whole concept of “fairness” and “equal access” as used by the regulators and rabble-rousers. Why are these things considered important points when discussing market structure? Are hospitals required to make operating rooms fairly accessible to DIY brain surgeons?

In his literature review, he (not surprisingly) refers to a number of papers I have discussed on PrefBlog before:

Biais and Green (2007) show that exchange-listed bond trading was quite liquid in municipal bonds before the late 1920s and in corporate bonds before the mid-1940s, and that transaction costs then were lower than they are now. The proliferation of electronic bond trading systems has the potential to substantially lower bond transaction costs, presumably to levels lower than Biais and Green document given the well-known economic efficiencies associated with electronic trading. Harris (2015) provides a survey of these efficiencies.

Well, that’s an inflammatory paragraph, isn’t it? But I reviewed Biais and Green in the post Exchange Traded Bonds? (emphasis added):

The third possibility [for the collapse of the exchange market] is due to the interaction of groups with differing objectives in a heterogeneous market:

Different equilibria will vary in terms of their attractiveness for different categories of market participants. Intermediaries benefit when liquidity concentrates in venues where they earn rents, such as opaque and fragmented markets. For reasons we will show were quite evident to observers at the time, large institutional investors fare better than retail investors in a dealership market. This was especially true on the NYSE until 1975, because commissions were regulated by the Constitution of the Exchange, while intermediary compensation was fully negotiable on the OTC market. We find that liquidity migrated from the exchange to the OTC market at times when institutional investors and dealers became more important relative to retail investors. As institutions and dealers became more prevalent in bond trading, they tipped the balance in favor of the over-the-counter markets.

Unlike many writers on this topic, Biais and Green show some understanding of the competing interests that determine market microstructure:

More Biais & Green:

Furthermore, the professionalized management and relatively frequent presence in the market of institutions makes transparency less important to them than to less sophisticated small investors who trade infrequently. The repeated interaction that dealers and institutions have with each other renders them less vulnerable to the opportunities which a lack of transparency affords other participants to profit at their expense on a one-time basis. Smaller institutions and individuals, for the opposite reasons, will tend to fare better in an exchange-based trading regime. Indeed, the theoretical model of Bernhardt et al (2005) shows that, in a dealer market, large institutions will trade more frequently and in larger amounts than retail investors, and incur lower transactions costs.(footnote)

Footnote: Bernhardt et al (2005) also offer an interesting empirical illustration of these effects in the case of the London Stock Exchange.

…there was a dramatic increase in institutional ownership in corporate bonds between 1940 and 1960. In the 1940s the weight and importance of institutional investors in the bond market grew tremendously. These investors came to amount for the majority of the trading activity in the bond market. Naturally, they chose to direct their trades to the OTC market, where they could effectively exploit their bargaining power, without being hindered by reporting and price priority constraints, and where they could avoid the regulated commissions which prevailed on the Exchange. Thus, the liquidity of the corporate bond market migrated to the dealer market.

Having cited Biais and Green, we may assume that Dr. Harris is familiar with these details, but he has chosen to ignore them in his efforts to increase market regulation.

One important point that goes against the thrust of the paper is the fact that:

The quotes used in this study are not generally available to the public, though they are available to IB’s customers in real-time.
…
Zitzewitz (2010) identifies RPTs, which he calls “trade pairing,” in the TRACE data using similar methods to those presented in this study. He finds that RPTs are very common (46% of trades under $100,000) and that they are mostly small trades. These results are similar to those obtained in this study.
…
Interactive Brokers serves as an agency-only broker for its clients. To facilitate their bond trades, IB collects pre-trade quotes and indications from several electronic trading platforms that offer automated execution services. These bond market centers include BondDesk, BONDLARGE, Knight BondPoint, NYSE Arca Bonds, and Tradeweb, and a few other centers that specialize only in municipal bonds or treasuries.13 None of these platforms provides universal coverage of all bonds that trade in the U.S. corporate bond markets. IB presents the quoted prices and sizes to its customers in real-time just as it and other brokers do for stocks, options, and futures.
…
IB reported to me that during the week ended September 10, 2015, they obtained complete fills for about 83% of its customers’ marketable orders and that they did not receive any cancellations after filling. This statistic indicates that a substantial fraction of the quoted and indicated prices that IB records are actionable.

The fact that all these quotes are available to anybody who signs up with Interactive Brokers shows that no regulatory changes are necessary. Anybody who wants to access these electronic quotes can do so. I see no problem here.

Dr. Harris does acknowledge the differing sizes of the retail and institutional trades:

Practitioners and academics often label trades with par values of $100,000 or less as retail-size trades, and larger trades as institutional-size trades. Many trades are relatively small retail-size trades. During the Primary Period, 67.3% of the trades in the full sample are retail-size trades (Table 9). Retail-size trades represent a slightly larger fraction (69.7%) in the subset sample. The median par value size of the retail-size trades is $18,000 in both samples.

The median trade size for institutional-size trades is $500,000 in both samples. The percentages of trades reported with indicators for par value sizes of $1,000,000 (speculative grade bonds) and $5,000,000 (investment grade bonds) or more are 4.6% and 1.3% in the full sample and about the same in the subset sample. Assuming that the actual size of these trades is equal to their minimum possible sizes of $1,000,000 and $5,000,000, the truncated mean par value trade size for all institutional-size trades is $908K and $953K in the two samples.
…
Among trades of a given size class, interdealer trades represent the smallest percentage of the largest class—those trades marked 5MM+ (13.1%). Many of these large trades probably are agency trades in which broker-dealers, acting as brokers, intermediate trades between customer buyers and sellers. In contrast, interdealer trades account for 40.8% of retail-size trades. The results in Section 7 show that many of these trades are riskless principal trades.

Of particular interest is the discussion of Table 19:

Most (82.3%) of the customer trade throughs are retail-size trades (Table 19). The mean price improvement for these trades is -93 bp, nearly a 1% markup. These markups seem quite large for relatively easy-to-arrange trades that can be arranged electronically. The total trade-through value for the retail trades is $74M. The mean price dis-improvement is smaller for institutional trades that traded through. Although these institutional trades are much larger, the total trade-through value is relatively small because these trades outsize the quotes. The average ratio of quote size to trade size is only 1.2% for institutional size trades in comparison to 28% for retail-size trades.
….
Standing quote to trade size ratio is the ratio of the opposing side quote size to the trade size.

So if I’m reading this correctly, the average size of a “trade-through” trade is four times the size of the quote, even when we restrict the sampling to retail sized trades (which average $18M, remember!). So these are itsy-bitsy little quotes and the “markups” calculated with respect to trade-through value would seem to be more of a market-impact cost than an extortionate dealer mark-up.

The number reported in the Bloomberg article comes from the introduction:

I find that average transaction costs that customers incur when trading range between 84.5 bp for retail size trades (under $100,000 in par value) and 52.1 bp for larger trades. These costs are several times larger than costs for similar size trades in equity markets. Trades occurring in markets with two-sided quotes that have stood at least two seconds trade through 46.8% of those markets; 40.8% of these trade throughs appear to be riskless principal transactions—trades for which the dealer has no inventory risk exposure usually because the dealer simultaneously offsets a trade with a customer with an interdealer trade. RPT transactions account for more than 41.7% of all trades. Total transaction costs borne by customers in U.S. corporate bond markets for the year ended March 31, 2015 are at least $26B, of which about $0.5B is due to trade-through value. During this period, markups on customer RPTs transferred $667M to dealers.

OK, so now we get to the good part, which is Dr. Harris’ Section 10.1, Public Policy Recommendations:

Many reasons explain why transaction costs are higher in bond markets than in stock markets. The most common explanation is that so many different bond issues make matching buyers to sellers difficult. This explanation certainly is true for the inactively traded bonds, but many bonds trade as actively as do small- and some mid-cap stocks, and they would undoubtedly trade much more actively if transaction costs were lower. Customers would benefit if the 850 bonds that are quoted nearly continuously were traded in market structures more similar to equity markets than the current OTC markets.

The problem with this paragraph is that much of it has not been supported by prior argument. Which stocks trade about as actively as which bonds, and what is the bid-offer spread on these stocks? How much size is there in these markets? Let us turn briefly to a speech by SEC Commissioner Luis A. Aguilar titled The Need for Greater Secondary Market Liquidity for Small Businesses:

In addition, it’s been reported that venture exchanges—both here and abroad—have suffered from low liquidity and, at times, high volatility.[19] This means investors could lose a lot of money quickly, and could have trouble selling their shares in a downturn. The Commission should attempt to determine the underlying causes of these problems and how best to address them. In this regard, we may need to ask some difficult questions. For example, should venture exchanges be structured as dealer markets, rather than auction markets? Also, could venture exchanges enhance liquidity through batch auctions, rather than continuous trading? How can the Commission, consistent with the Exchange Act, encourage traders to execute transactions on venture exchanges, rather than in off-exchange venues?[20] And, finally, could larger ticker sizes enhance liquidity by encouraging market maker activity and fostering research coverage? In this regard, the Commission’s proposed tick size pilot program[21] may offer valuable insights on the role of tick sizes in ensuring an active secondary market for smaller companies.

So for at least some of these smaller issues there are musings about possibly moving the other way – from exchange trading to a dealer market! We can also look at the fascinating Table 2 from the SEC’s report A characterization of market quality for small capitalization US equities:

Click for Big

Given that most bond issues will be smaller than the largest equities on this table and that bonds trade less intensively than equities, it is clear that that a Dr. Harris needs to support his recommendation in considerably more detail than he has in this paper. As an aside, I think he needs to explain his tables a little better! I can’t figure out what 850 bonds he’s talking about in his Table 11! And when he’s talking about “size”, I don’t know if he’s talking about dollar value, multiples of 100, or multiples of 1000 … I suspect he means multiples of 100, but it’s certainly not very clear!

Another recommendation is:

The SEC also should consider enacting a trade through rule for bonds similar to that in Reg NMS (for equities) that would require that broker-dealers access electronically available orders when filling orders for their clients before trading through. The SEC may want to do so before a class-action lawsuit based on common law agency principles effectively imposes a Manning Rule for bonds similar to FINRA Rule 5320 (Prohibition Against Trading Ahead of Customer Orders) for equities

Well, I’m not going to pretend to know anything about the Manning Rule, or the chances that a class-action lawsuit might have! However, I will point out that while this might well apply to dealers acting as brokers, it does not apply to dealers acting as principal.

Dr. Harris suggests:

At the minimum, FINRA or the SEC should require that brokers disclose their markup rates on RPTs on a pre-trade basis as they do with their commission rates.34 Since the two rates are perfect substitutes for each other, investors would be less confused if one rate were simply set to zero. This brokerage pricing standard would ensure that brokers would compete on the same basis for order flow. Since customers understand commissions much better than they understand markups, simply banning markups on RPTs would be best. Such a ban would have no effect on competition because dealers could always raise their commissions to compensate for their lost markups. Their customers then would know the full cost of the intermediation services that they obtain from their brokers.

Readers who have gotten this far will know that I am going to object to the assertion that commissions and markups are equivalent – the former applies to brokerage and the latter to principals. I will also note that while full-service commissions are highly variable and considered top-secret, 1% for equity trades is a good place to start. The comparisons here appear to be with equity transactions via a discount brokerage, which is a different kettle of fish.

Update, 2015-9-27: I note from Rob Carrick’s fee project:

Now back to Dr. Harris:

Finally, a rule that would require brokers to post limit orders of willing customers to venues (order display facilities) that widely disseminate these prices would help prevent many trade throughs. Many trade throughs undoubtedly happen simply because traders are unaware of better prices. Such a rule likely would substantially increase such offers of liquidity, especially if implemented in conjunction with a trade-through rule. These order display facilities could be existing exchanges and ATSs, or new ones formed for this purpose.

This follows from the idea that bond dealers act as brokers and the marketplace is an exchange.

If the SEC fails to take these actions, and if no class-action suit is successful, the markets will continue to improve as innovators such as IB continue to capture order flow by creating their own NBBOs. But it may be many years before most customers become sophisticated enough to demand these facilities from their brokers, if they ever do, and some brokers may never offer these facilities, either because their customers are not well enough informed or because their customers suffer various agency problems, including the problems associated with payments for order flow.

It’s the profitability of ‘innovators such as IB’ that makes the debate unnecessary. Let competition reign – particularly since for small investors the real competition is ETFs and funds.

With respect to trading, bonds are securities just like equities, only less risky. U.S. corporate and municipal bonds presently trade differently for historic reasons. They need not trade differently in the future. U.S. Treasury bonds and corporate bonds in several well developed countries trade in substantially more transparent markets that do corporate and municipal bonds in the U.S. presently do. The quality of these markets shows that opaque markets are not necessary for fixed income securities.

This paragraph is not supported by the text and ignores the work that has been done on market microstructure as it relates to market-depth and transparency.

Finally, note that the creation of more liquid markets will benefit issuers as well as customers. Investors are more willing to buy securities in the primary markets when they expect that they can sell them easily at low cost in the secondary markets. Low secondary trading costs thus imply higher bond IPO values, and lower corporate funding costs.

While it’s nice to see a nod to the interests of issuers, the evidence actually goes the other way. In the Bessembinder paper I reviewed in the post TRACE and Corporate Bond Market Transparency, it is shown that increased transparency caused a migration to less transparent “144a” structures, which are private placements:

One way to circumvent TRACE, which applies to publicly-issued bonds, is for a firm to issue privately placed bonds (sometimes referred to as Rule 144a securities, for the section of the Securities Act of 1933 that provides exemption from registration requirements). … In 2001, before TRACE, “144a for life” bonds were 7.3 percent of dollar volume and 9.6 percent of issues. The percentage of dollar volume in “144a for life” bonds jumped to 27.8 percent in 2003, the first full year after TRACE initiation, and grew to 39.8 percent in 2004, before declining to 16.9 percent in 2006.…
Also consistent with a shift towards alternative asset classes, the credit default swap market experienced phenomenal growth in recent years relative to bonds. Table 6 reports on outstanding notional principal in these credit default swaps, which grew from $919 billion in 2001 to $34.4 trillion in 2006. One dealer suggested to us that, prior to TRACE introduction, ten times as much capital was allocated to corporate bond trading than to credit default swaps, but that the ratio has now been reversed.
…
To the extent that the shift to privately placed bonds and bank loans was initiated by corporate borrowers, and in response to TRACE, it suggests that the net costs of TRACE may exceed the benefits….

All in all, it’s an interesting paper and a good reminder that corporate bond trades should ensure they have independent access to electronic marketplaces … but note that if a dealer is sitting on a stack of inventory he’s willing to sell at 102.00, then sees all the offers below 102.00 disappear, he’s probably going to raise his price! But the data needs to be presented with more explanation in the tables and the advocacy should be taken out and used elsewhere; in addition, more account needs to be taken of previous work on market microstructure and the interests of issuers which, I assert, must be paramount when contemplating changes to the system.

Assiduous Readers will remember that I was recently quoted as saying:

“It is simply not sustainable for a five-year Canada to trade below inflation forever,” he said. “That simply cannot go on.”

I have been challenged to substantiate this assertion and my immediate response was:

You are correct – it’s because the real yield is not just negative, but significantly negative.

Some bond investors have to put up with this kind of thing; banks, for instance, are required to hold a large quantity of government bonds. Central banks will, in general, place a very high premium on liquidity, since when they want to trade they want do it in size and they also place a high premium on the ability to transact at the height of a crisis.

But the marginal investor will eventually get tired of losing money on a real basis while tying up their money for five years. In addition, the marginal borrower will step up borrowing because he’s getting paid – in real terms – to do so. We are already seeing significant economic distortions resulting from these marginal borrowers because instead of buying productive assets, they’re buying houses in Toronto and Vancouver … eventually, all the stresses will be relieved, and there will be a positive real yield on five year Canadas … either because government yields go up or because we enter a period of deflation.

These will be familiar themes to Assiduous Readers; the requirement for banks to own an increasing number of sovereign bonds for highly politicized (and economically illiterate and fiscally expedient) reasons was discussed on September 4, for instance.

While poking around for more authoritative and crushing retorts, I came across Remarks by Hervé Hannoun, Deputy General Manager, Bank for International Settlements, at the Eurofi High-Level Seminar, Riga, 22 April 2015, titled Ultra-low or negative interest rates: what they mean for financial stability and growth. He was mainly concerned with the European situation:

When policy interest rates came down to almost zero and central bank balance sheets expanded due to large-scale market interventions in the wake of the Global Financial Crisis, the consensus was that this unconventional monetary policy (UMP) would be temporary. More than six years later, the prospect of normalisation seems remote in most advanced economies. Indeed, most of continental Europe (the euro zone, Denmark, Sweden and Switzerland) have moved towards a much more extreme form of UMP by introducing negative policy interest rates, and/or negative central bank deposit rates. Together with forward guidance and large scale asset purchases, such measures have created an unprecedented situation where nominal interest rates in a number of European countries are negative across a range of maturities in the benchmark yield curve, from overnight out to five years.
…
There is no precedent in economic history for negative nominal interest rates, even during the Great Depression in the United States.[Footnote] Not even Keynes, who coined the terrifying metaphor of the “euthanasia of the rentiers”, ever contemplated negative nominal interest rates. An experiment is under way in continental Europe to test the “boundaries of the unthinkable” in monetary policy.

[Footnote reads] In the wake of the Great Depression, US short-term nominal interest rates fell to near-zero levels in 1932 but they never turned negative.

He shouldn’t be quite so absolute in his footnotes! There were brief episodes of negative US bill rates in 2013:

Treasury bills that mature as soon as November traded below zero today [2013-9-26], with the bill maturing on Nov. 29 having a negative 0.005 percent rate at 2:02 p.m. New York time. The three-month bill rate was negative 0.0051 percent, compared to 0.0152 percent yesterday. Treasury bills that mature on Oct. 24 were at a rate of 0.038 percent, up from 0.018 percent yesterday.

and in 2014:

A scramble for safe, short-term debt left some investors on Tuesday paying for the privilege of lending to the U.S. government.

The demand, which intensified following the Federal Reserve’s decision this month to curb a popular overnight-lending program, pushed up bond prices and drove down yields. The yield on the U.S. Treasury bill maturing on Oct. 2 traded at negative-0.01%, according to Tradeweb, the first negative yield in eight months. Yields on other Treasury bills due in three months or less hovered around zero.

Short-term debt trading at negative yields was essentially unheard of before the 2008 financial crisis. But since then, the condition has cropped up at times of market stress, reflecting extraordinarily expansive central-bank policy and anemic growth in much of the world. Yields on some U.S. bills traded below zero at the end of each of the past three years amid strong demand for liquid assets, according to analysts.

… and shortly after the speech:

For all the anxiety over the global selloff in bonds, the big worry in money markets is the havoc being created by a dearth of U.S. Treasury bills.

The magnitude of the problem was on display last week, when not even the Treasury Department’s surprise announcement to boost sales could do much to lift bill rates. Over the past two weeks, some of those rates have turned negative, reaching levels last seen during the financial crisis.

With supply at multi-decade lows, investors are signaling alarm as regulations intended to shore up banks and prevent a run on money-market funds exacerbate the bill shortfall. JPMorgan Chase & Co. expects an extra $900 billion of demand for government securities during the next 18 months, putting pressure on a sizable chunk of the $1.4 trillion bill market.
…
The mismatch between supply and demand has been so acute that four-week bill rates fell to minus 0.0304 percent on April 29, the lowest on a closing basis since December 2008. Yields on three-month bills also turned negative. The Treasury responded by saying at its quarterly refunding announcement on May 6 it would increase issuance to meet growing demand.

… and in the US in the Great Depression:

1The interest rate on Treasury bills would tend to not fall below zero if currency incurs no taxes, storage costs, or insurance costs. Absent such costs, if the Treasury bill rate were to be negative the holders of Treasury bills would prefer to hold currency because currency has the advantage of being a more liquid asset and its implicit interest rate of zero would be greater than the negative rate on Treasury bills. Holders of Treasury bills would sell them, drive down their price, and increase their interest rate until the interest rate reaches at least zero. For simplicity, we assume the lower bound on short-term interest rates is zero, even though nominal yields dropped slightly below zero in the United States in the Great Depression (as discussed in footnote 15) and in Japan recently (as discussed in footnote 22).
…
15See Federal Reserve Board (1943, p. 462). Two reasons for this phenomenon, perhaps responsible for a few basis points of the negative yield on Treasury bills, are as follows: First, Treasury bills were exempt from personal property taxes in some states, while cash was not. Thus, the after-tax rate of return on cash holdings was negative in some states. Second, Treasury securities were required as collateral for a bank to hold U.S. government deposits, so the total return, net of the collateral benefits, could have been zero or positive for banks. During this period, negative yields were also reported on Treasury bonds with up to two years maturity, owing to a valuable exchange privilege implicit in holding the securities. Cecchetti (1988) provides a detailed explanation of this phenomenon and shows that once the value of this exchange privilege is accounted for, yield estimates on those securities become positive. These factors allowing for negative pecuniary yields emphasize that institutional considerations such as these make it dicult to be precise about the actual lower bound for nominal interest rates.

But let’s be fair … perhaps Mr. Hannoun meant “short-term bonds” when he said “short-term” and perhaps he is not so much of a pedant as to specify “after accounting for special privileges”. Still, he could quite easily have said “rare and transient”.

At any rate, Mr. Hannoun first reviewed the effects of low interest rates on growth:

In essence, the monetary stimulus aims to lift short-term growth via five main channels: by boosting credit to the real economy (the credit channel), by lifting asset prices (the asset valuation channel), by forcing investors away from safe assets towards riskier ones (the portfolio balance and risktaking channels), by lowering the exchange rate (the exchange rate channel) and by attempting to nudge inflation up towards objectives with a view to warding off a so-called deflationary spiral (the reflation channel).

It’s all good stuff, but one thing worth highlighting is his discussion of the portfolio balance and risk-taking channel:

Advocates of UMP argue that these policies will encourage investors to shift out of government bonds and into riskier assets. This is the portfolio balance channel. Indeed, the search for yield engineered by zero or negative nominal policy interest rates has fuelled more risk-taking, leading to a convergence between the returns of risky assets and those of low-risk assets, as currently seen in the euro zone’s sovereign credit spreads. These appear to be re-enacting the extreme compression of sovereign spreads, invariant to differences in credit quality, that occurred before the crisis (Graph 2). If, as many would agree, the euro zone’s sovereign risks were mispriced then, we now seem to be heading back to that situation. The European Commission’s prudential policy of applying a uniform zero risk weight to all sovereigns in EU bank regulation strengthens this effect. The problem here is that risk weights are not differentiated according to credit quality, contrary to the Basel II requirements.
…
In reaction to negative yields in the short- and medium-term segment of the euro zone sovereign yield curves, investors are piling up interest rate risk by investing in long-dated securities at very low yields. And, in fact, the effective duration of euro-denominated debt has risen significantly since the second half of 2014 (Graph 3). As a result, an eventual normalisation of long-term yields would inflict significant and widespread losses on investors, with potentially serious consequences for financial and economic stability.

This makes it a matter of urgency to address the gap in global regulation on interest rate risk in the banking book. Pillar 1 currently does not provide for any capital charge against this risk, an anomaly that will, we hope, soon be corrected by the Basel Committee on Banking Supervision (BCBS). As monetary policymakers, central banks in Europe have contributed heavily to the build-up of duration risk by bringing nominal yields in the two- to five-year part of the yield curve down to near zero and even negative levels. As supervisors or systemic risk managers, they should ensure that commercial banks are allocating enough capital to cover the interest rate risk they are accumulating. All this puts a premium on introducing a Pillar 1 charge on interest rate risk in the banking book as soon as possible.

Click for Big

However, I’m more interested in the section titled Longer-term unintended consequences of ultra-low or negative
interest rates (emphasis added):

From a longer-term perspective, there are five main risks that may make the prolongation of ultra-low or negative nominal interest rates counterproductive. These can be summarised as: disincentive, distraction, distortion, disruption and disillusion.
…
Bond market prices in the euro zone may no longer adequately reflect the risk inherent in record high debt levels. At the same time, equity prices are artificially inflated as investors are forced into increasingly risky assets. All this involves the risk of a major correction when confidence in inflated valuations is lost. The question is not whether this will happen again, but when. Of course, nobody can say when the next “Minsky moment”, a generalised loss of confidence in artificially inflated valuations, will occur. Yet there is no doubt that the probability and severity of another financial crisis is increased by the prolongation of ultra-low or negative rates.

Advocates of ultra-low or negative interest rates argue that macroprudential tools can be used to offset/mitigate the financial risks and distortions resulting from ultra-easy monetary policy.
…
The risk of disillusion also looms large for households in the new world of low returns. Ultra-low or negative interest rates will add to their worries by making it difficult for them to build up enough retirement savings. Thus households are more likely to increase their savings rate than to reduce it.

Negative rates on deposited savings – effectively a form of taxation – will feed the debate on the
“financial repression” of savers.

True, some households stand to gain from low mortgage rates, but this benefit will accrue only to those who can afford to buy a house. Moreover, the positive effect of low mortgage loan rates is largely offset by the increase in property prices fuelled by ultra-low interest rates.

There is also the question of inequality. Most households will lose from the dwindling returns on their savings without gaining anything from asset price inflation. They are not sophisticated asset managers who can realise capital gains in financial markets when long-term yields fall, and they will be affected by the low returns on their savings.

It will be noted that the word “macroprudential” is basically the new cool way to say “credit rationing”:

In the UK the Bank of England has been flashing an amber light for months about the complacency shown by low market volatility, but in house-price obsessed Britain, mortgage excess is the focus of its worry. Last month it became the first of the major central banks to set out to try to control credit using non-monetary tools: in the jargon, “macroprudential measures”. Ms Yellen has been highlighting macropru as the first line of defence against bubbles for a while.

The problem is simple. Central bankers want money to lubricate the real economy, not to flow into pointless leverage of existing assets. Higher rates could reduce the incentives to leverage, but at the cost of damage to the real economy. Their solution is to set up barriers inside the banks to direct the flow.
…
As financial historian and CLSA consultant Russell Napier pointed out recently, credit rationing was a disaster in the 1970s. The theory relies on markets being so bad at allocating resources that the job is better done by a handful of men and women at central banks and regulators.

We’ve seen some of this credit rationing in Canada, of course, in the tightening of mortgage requirements by the CMHC and bank regulators; wildly cheered on by established Canadians who inherited their house from Mommy and Daddy and don’t like the idea that immigrants and the working class might get a shot at ownership.

Mr. Hannoun concludes:

The policy of prolonged ultra-low, or negative, interest rates relies on transmission channels with uncertain effectiveness and potentially serious unintended consequences. For central banks, such policies raise the risk of financial dominance, exchange rate dominance and fiscal dominance – that is, the danger that monetary policy becomes subordinated to the demands of propping up financial markets, massaging the exchange rate downwards, and keeping public refinancing costs low in the face of unprecedented public debt burdens. These risks have been present before,13 but never so acutely as today.

Meanwhile, financial markets continue to set the stage for policy deliberations by fuelling expectations for continued, and additional, monetary accommodation. Behind the enthusiasm of market participants for extreme monetary policy, of course, lurks the fear that asset prices might collapse when the music of monetary easing stops.

So, yeah, this speech belongs in a list of justifications for my statement!