Summary Fixed Income
Week 2: Guest Lecture on Expected Issuance Fees and Market Liquidity
§ Summary of the paper
- The paper examines the interaction between the primary and secondary markets for Euro area
sovereign bonds.
® Primary market is the market for new issuances.
Secondary market is where the trading of the bond takes place.
® Primary dealers determine how the bonds are being placed in the primary market and the rules for
trading in the secondary market.
- Sample: sovereign bonds from 11 Euro area countries from MTS from 2008-2011.
® Most of the trading in the secondary market is done OTC. For eurozone government bonds the
trading happens through electronic platform MTS (an exchange like market).
- Result: fee-driven liquidity effect
The expected issuance fees are positively related to market liquidity.
® Explanation: The Debt Management Office (DMO) places the new issues into the new market. They
do this through the primary market via syndicated issues. These syndicated issues earn a fee. Banks
compete to get this mandate (so that they can place the issue in the market). The DMO tells them that
they have a higher chance of getting the mandate if they provide liquidity on the MTS platform.
® The fee-driven liquidity is especially strong for countries with high funding needs (Italy), in periods
of high re-financing uncertainty, and for low-risk bonds.
§ Key player 1: The Issuer
- The issuer is the sovereign or in effect the DMO.
- Bonds have a base interest rate component, a credit risk and liquidity risk component.
The higher the credit risk à the higher the premium à the higher yield of the bond
The lower liquidity of the bond à the higher the illiquidity premium à higher yield of the bond
® DMO can’t influence the base interest rates (done by central bank) and credit risk component
(different part of the ministry of finance). They might be able to affect the liquidity component.
- Issuer has an incentive to minimize the bond yield (is lower when the bond liquidity), because then
they obtain a higher price.
§ Key player 2: Banks
- Investment banks are appointed as primary dealers by the DMOs. They operate as underwriter in the
primary market and as market makers in the secondary market to provide liquidity.
® Pay-offs primary market: Issuance Fees
® Pay-offs secondary market: Bid-ask spread
- Competition for issuance fees among banks
® Issuer selects (lead) underwriter for new bond issues based on liquidity provision.
® Very competitive as market making is a loss-making business and underwriting earns fee.
® Banks have an incentive to post extra liquidity in order to have a higher probability to be selected
as underwriter. à Hypothesis: higher expected issuance fees leads to higher liquidity.
,§ Institutional setting
- The European sovereign bond market provides a very good test case, because:
a) Lot of heterogeneity between the different sovereign markets, but all countries share the same
currency.
b) The only market which has an exchange-like marketplace for sovereign bonds
§ Related literature
- In market liquidity studies, risk-related factors are pointed as the most influencing factors:
1) Costs of carrying inventory (= the number of bonds that you want to hold as a trading desk). The
more inventory you have, the more immediate you can provide the customer with the bond, so
you want to pass the costs of having inventory on.
2) Adverse selection costs (= the costs of having to deal with investors better informed than you).
Adverse selection costs reduce liquidity.
3) Order processing costs. The higher the costs of processing orders, the lower the liquidity.
® We extend this by adding the issuance fee-driven liquidity mechanism.
§ MTS
- MTS is an electronic and central platform through which the various DMOs in Europe promote and
measure liquidity provision by their primary dealers.
- MTS is very relevant, because of:
a) The amount of trading
b) Most dominant price setter in the EZ government bond market
§ Data: liquidity
- Ultra-high frequency data from MTS àData on bid and ask prices, trade pricing and volumes.
- What is market liquidity? Ease by which one can liquidate or deliquiate assets.
- Liquidity measure?
1. Price liquidity measure: take the best bid and best ask prices
If the bid-ask price is wide à liquidity is poor
If the bid-ask price is narrow à liquidity is good
2. Volume based measure: look at the total volume that is quoted and ignore the prices.
3. Limit-Orderbook-Slope (LOS): combines the price and volume information
® A line from the three best bid and three best ask prices with the cumulative volume and then
you get a straight line.
®The lower the slope, the larger the liquidity (for a large volume change, smaller price change).
§ Data: Issuance fees
- Paper obtains data on issuance fees from deallogic.
® When new bonds came to the primary market, how much fee was attached and who was the lead
underwriter.
- Issuance fee measure (IFM): a backward-looking expected issuance fee is constructed.
- There is a lot of variation in the fees that you can expect between sovereigns.
® Issuance fees are not there for Germany and the Netherlands, because they don’t issue through
syndicates but exclusively through auctions.
® Issuance fees for Italy are very high and frequent.
,§ Empirical analysis
- We regress the Liquidity Measure on the Issuance Measure and control variables:
L: Liquidity measure (slope)
I: Issuance measure
X: Control variables (proved to have impact on liquidity)
• Bond level-variables (maturity, seasonedness, credit spread, issue size)
• Funding liquidity: amount of stress in the marketplace
• Volatility: volatility in the respected equity markets of each sovereign.
• #dealers
• Traded volume
µ: Time fixed effects
n: Country fixed effects
- The paper looks whether there is an additional effect, which is the Issuance Measure.
§ Main result
- Issuance fee measure has significantly positive effect: the higher the issuance fee, the higher liquidity.
- All control variables have the expected sign:
• The longer the maturity, the lower the amount of liquidity
• The higher the volatility, the lower the amount of liquidity
• The lower the funding liquidity (higher amount of stress), the lower the amount of liquidity.
• With a larger traded volume, you have better liquidity.
- In addition to the above results, where every variable is added separately, they look at interactions.
o Issuer interactions
® We expect the fee-driven liquidity effect to be increasing in volatility, decreasing in funding
liquidity, decreasing during the LTRO period, and increasing in the subset of peripheral countries.
® Factor Uncertainty is the set consisting of Funding liquidity, Volatility, LTRO, and Peripheral:
Estimation of the interaction effects (ß2) is significant and has the sign we expect.
o Market maker interactions
® This set has to do with bond characteristics
® We expect that the primary dealer will increase the liquidity of those bonds that are safest, and
that the existence of a derivative market makes liquidity providing less costly (liquidity is less
important for these sovereigns).
® Factor Bond is the set consisting of Derivatives, Maturity, Seasonedness, Issue size, Traded
volume, and Credit spread. We add this to the equation:
ß2 is statistically significant and has the sign we expect.
® Derivatives are futures. There are three markets that have very active future markets:
Germany, Italy and France.
o Number of Dealers interaction
® We expect that an increase in the number of primary dealers strengthens the fee-driven
mechanism (more competition and therefore liquidity is greater).
, ß2 is statistically significant and has the sign we expect.
- Full model test: all interactions remain significant when included simultaneously.
- Our results are robust for:
1. Alternative liquidity measures (bid-ask measure, volume measure)
2. Alternative issuance fee measures (forward-looking fee measure)
3. To some other empirical choices (set of countries, positive trading volume, and bond fixed effects)
§ Conclusion
- Expected issuance fee is significantly related to government bond market liquidity (they controlled for
risk-related factors).
- Issuance fee-driven liquidity mechanism is proven to exist à liquidity does not only depend on the
common factors such as inventory risk and adverse selection, but also on the extent to which the
sovereign is willing to compensate the primary dealer for their services.
- The issuance fee-driven liquidity mechanism is especially strong for countries with relatively high
funding needs, in periods of high market uncertainty, and for safe bonds.
Lecture 2: Interest Rate Risk
§ Introduction
- The different sectors of the bond markets and their relevant risks:
1. Treasury sector (‘govies’) Interest rate and reinvestment risks
® Treasuries determine the basic level of interest. Therefore, also the reinvestment risks.
® US treasury bonds and German Government bonds are the best proxy for the risk-free level of
interest rate, because the US and German government are very creditworthy.
2. Corporate bond sector (‘eurobonds’) + Credit Risk
The level of credit risk rises in the corporate bond sector.
3. Mortgage sector (‘MBS’) + Call Risk (prepayment risk)
4. Non-US bond market + Exchange-rate risk
- This lecture discusses measures to determine the future level of interest rates.
® One way is forward rates. Forward rates are what the market implies the future levels of interest
rates to be. Downsides: 1) Embedded in the current yield curve and when the yield curve changes the
forward rate changes (yield curve changes every second), and 2) shown that forward rates are not
good at predicting future level of interest rates.
§ Static and Option Adjusted Spread
- Option Adjusted Spread (OAS): yield spread between two bonds after adjusting for the value of
embedded options.
® The spread is option adjusted because it allows for future interest rate volatility to affect the cash
flows.
- Traditional yield spread approach: difference in yield between a bond not on the yield curve and the
yield curve.
® Static spread fails to take the following factors into account:
1) The term structure of interest rates (assumes that interest rates will not change)
2) The options embedded in the bond
3) The expected volatility of interest rates
