Ruslan Goyenko

Ruslan Goyenko

Treasury Illiquidity and Funding Liquidity Risk

Treasury Illiquidity and Funding Liquidity Risk

PRMIA-CIRANO- Lunch Conference December 7, 2011

Ruslan GoyenkoMcGill University

ILLIQUIDITY

MARKET ILLIQUIDITY FUNDING LIQUIDITY

We know how to measure How do we measure it?

It is priced Is it priced?

What the paper does:

Christoffersen, Goyenko, Jacobs, and Karoui Illiquidity Premia in the Equity Options MarketRuslan Goyenko Treasury Illiquidity and Funding Liquidity Risk

Treasury Illiquidity and Funding Liquidity


Funding Problems

Reduced positions

Higher margins

Price moves away from

fundamentals

Losses on existing

positions

Outflow from the stock market

Treasury Illiquidity


Market Liquidity – reflects the ease of trading in bond markets (Goyenko, Subrahmanyam and Ukhov, JFQA, 2011)

Also predicts the ease of trading in the stock market (Goyenko and Ukhov, JFQA, 2009)

What else?

Treasury Illiquidity


TED SPREAD


Treasury Illiquidity and Funding Liquidity


1999/1 2000/1 2001/1 2002/1 2003/1 2004/1 2005/1 2006/1 2007/1 2008/1 2009/10

0.001

0.002

0.003

0.004

0.005

0.006

0.007

0.008

Treasury ILLIQ

1999/1 2000/1 2001/1 2002/1 2003/1 2004/1 2005/1 2006/1 2007/1 2008/1 2009/10

0.0005

0.001

0.0015

0.002

0.0025

0.003

0.0035

0.004

0.0045

TED

For the sample period 01/1971 to 12/2009, correlation=0.78

VAR(2): FED, Stock Illiquidity, Bond Illiquidity and TED spread


-.0001

.0000

.0001

.0002

.0003

.0004

1 2 3 4 5 6 7 8 9 10

Response of TED SPREAD to FED

-.0001

.0000

.0001

.0002

.0003

.0004

1 2 3 4 5 6 7 8 9 10

Response of TED SPREAD to Stock ILLIQ

-.0001

.0000

.0001

.0002

.0003

.0004

1 2 3 4 5 6 7 8 9 10

Response of TED SPREAD to Bond ILLIQ

-.0001

.0000

.0001

.0002

.0003

.0004

1 2 3 4 5 6 7 8 9 10

Response of TED SPREAD to TED SPREAD

Response to Cholesky One S.D. Innovations ± 2 S.E.

VAR(2): FED, Stock Illiquidity, Bond Illiquidity and TED spread


-0.5

0.0

0.5

1.0

1.5

2.0

1 2 3 4 5 6 7 8 9 10

Response of Stock ILLIQ to FED

-0.5

0.0

0.5

1.0

1.5

2.0

1 2 3 4 5 6 7 8 9 10

Response of Stock ILLIQ to Bond ILLIQ

-0.5

0.0

0.5

1.0

1.5

2.0

1 2 3 4 5 6 7 8 9 10

Response of Stock ILLIQ to TED SPREAD

-.002

-.001

.000

.001

.002

.003

.004

1 2 3 4 5 6 7 8 9 10

Response of Bond ILLIQ to FED

-.002

-.001

.000

.001

.002

.003

.004

1 2 3 4 5 6 7 8 9 10

Response of Bond ILLIQ to Stock ILLIQ

-.002

-.001

.000

.001

.002

.003

.004

1 2 3 4 5 6 7 8 9 10

Response of Bond ILLIQ to TED SPREAD

Response to Cholesky One S.D. Innovations ± 2 S.E.

Bond-ILLIQ beta portfolios, from 5-factor model


VW-Ret t-stat FFC-Alpha t-stat

Low Portf 1 0.203 0.59 -0.363 -2.032 0.494 1.67 0.020 0.143 0.512 1.93 0.123 0.954 0.374 1.49 -0.032 -0.265 0.468 1.90 -0.010 -0.096 0.396 1.69 -0.079 -0.677 0.378 1.68 -0.092 -0.918 0.247 1.15 -0.228 -2.309 0.563 2.69 0.170 1.60

10 0.505 2.41 0.025 0.2711 0.446 2.17 0.010 0.1012 0.470 2.29 -0.010 -0.1113 0.496 2.31 0.041 0.4214 0.556 2.54 0.004 0.0415 0.596 2.79 0.056 0.5116 0.496 2.21 -0.001 -0.0117 0.578 2.45 -0.068 -0.6118 0.668 2.82 0.065 0.5319 0.538 2.05 -0.110 -0.79

High Portf 20 0.719 2.25 0.128 0.71

High-Low 0.517 2.11 0.491 1.97

Bond-ILLIQ beta portfolios, from 5-factor model


1971-1990 1991-2009

VW-Ret t-statFFC-Alpha t-stat VW-Ret t-stat

FFC-Alpha t-stat

Low Portf 1 -0.025 -0.05 -0.343 -1.52 Portf 1 0.420 0.85 -0.343 -1.22

2 0.414 1.01 0.005 0.03 2 0.549 1.27 -0.013 -0.06

3 0.454 1.19 0.271 1.54 3 0.540 1.46 0.032 0.17

4 0.362 1.01 0.097 0.59 4 0.342 0.97 -0.167 -0.87

5 0.367 1.04 0.002 0.02 5 0.551 1.61 0.017 0.10

6 0.427 1.34 0.112 0.79 6 0.348 1.01 -0.237 -1.23

7 0.313 0.94 0.087 0.68 7 0.426 1.40 -0.154 -0.99

8 0.159 0.49 -0.195 -1.47 8 0.311 1.11 -0.182 -1.27

9 0.505 1.58 0.265 2.01 9 0.599 2.24 0.192 1.22

10 0.401 1.26 0.139 1.17 10 0.582 2.13 0.040 0.30

11 0.309 0.98 0.020 0.18 11 0.565 2.19 0.138 0.98

12 0.391 1.31 -0.013 -0.12 12 0.537 1.89 0.037 0.25

13 0.398 1.25 -0.015 -0.13 13 0.580 2.00 0.141 0.92

14 0.577 1.78 0.156 1.31 14 0.507 1.72 -0.076 -0.46

15 0.600 1.86 0.148 1.21 15 0.583 2.09 0.029 0.17

16 0.456 1.33 0.096 0.74 16 0.508 1.77 0.020 0.12

17 0.484 1.41 0.005 0.04 17 0.647 2.01 -0.077 -0.45

18 0.536 1.51 0.061 0.39 18 0.784 2.50 0.143 0.78

19 0.390 1.01 -0.109 -0.66 19 0.674 1.90 -0.054 -0.25

HighPortf

200.379 0.82 -0.205 -0.96

Portf 20

1.072 2.42 0.439 1.57

High-Low 0.403 1.29 0.138 0.43 High-Low 0.652 1.71 0.782 2.01

Size-Bond ILLIQ Beta quintile portfolios (1/1971 to 12/2009)


FFC-Alpha t-statistics

Size 1 2 3 4 Size 5Small-

Big Size 1 2 3 4 Size 5Small-

Big

Beta 1 -0.283 -0.192 -0.168 -0.004 0.034 -0.317 Beta 1 -2.16 -1.72 -1.50 -0.04 0.40 -2.03

2 0.118 -0.007 0.127 0.077 -0.111 0.229 2 0.98 -0.07 1.37 0.86 -1.57 1.76

3 -0.018 0.029 0.180 0.050 0.000 -0.017 3 -0.16 0.30 1.94 0.59 -0.01 -0.14

4 -0.031 -0.001 0.092 0.003 0.010 -0.041 4 -0.27 -0.01 0.98 0.03 0.13 -0.34

Beta 5 0.066 -0.090 0.102 -0.056 0.006 0.061 Beta 5 0.46 -0.72 0.80 -0.52 0.06 0.4

High-Low 0.350 0.102 0.270 -0.053 -0.028

High-Low 2.23 0.71 2.11 -0.41 -0.18

Amihud ILLIQ-Bond ILLIQ Beta quintile portfolios (1/1971 to 12/2009)



Low-ILLIQ 2 3 4

High-ILLIQ

High-Low

Low-ILLIQ 2 3 4

High-ILLIQ

High-Low

Beta 1 0.054 -0.080 -0.038 -0.113 -0.321 -0.375 Beta 1 0.60 -0.73 -0.34 -1.03 -2.31 -2.22

2 -0.136 0.117 0.047 0.099 0.240 0.376 2 -1.93 1.33 0.50 0.98 2.13 3.08

3 -0.012 0.064 0.179 0.140 0.042 0.054 3 -0.17 0.70 1.92 1.40 0.38 0.44

4 -0.009 0.027 0.081 0.102 -0.155 -0.146 4 -0.11 0.30 0.81 1.03 -1.38 -1.18

Beta 5 -0.014 0.017 0.027 0.060 0.170 0.184 Beta 5 -0.13 0.16 0.23 0.50 1.23 1.23

High-Low -0.068 0.097 0.066 0.173 0.491

High-Low -0.43 0.69 0.49 1.24 2.95

The model

The loadings are pre-estimated from portfolios 10sizex10bond-beta (similar to Fama-French 1992)

Cross-Sectional Fama-MacBeth Regressions


tititti ZRE ,1,0

titiiti fR ,,0,

Risk Premiums


Firm Characteristics

MKT SMB HML UMD Bond-ILLIQ

Sizet-1 Amihudt-1 B/Mt-12

-0.003(0.76)

0.004(2.78)

-0.003(0.95)

0.001(0.46)

0.004(1.84)

0.004(2.74)

-0.002(0.54)

0.001(0.38)

0.005(2.02)

0.006(1.54)

0.003(2.28)

-0.003(1.04)

0.003(1.26)

0.001(0.67)

0.006(1.45)

0.003(2.06)

-2.7e-11(0.57)

-0.001(2.44)

0.002(3.81)

Asset fire sales (Coval and Stafford, JFE, 2007) – fund manager needs to liquidate at below fundamental values when a fund underperforms and investors withdraw

Capital-constrained assets are even more difficult to liquidate

The main hypothesis: This extra risk of holding capital-constrained assets should have extra reward

Active Equity Mutual Funds and Funding Liquidity


Active Equity Mutual Funds – 2,546 funds, 1/1990 to 12/2009


1[Low]

2 3 4 5 6 7 8 9 10[High]

High-Low

ExcessReturn

-0.420(1.88)

-0.286(2.32)

-0.219(2.42)

-0.191(3.30)

-0.190(4.05)

-0.126(2.90)

-0.066(1.29)

0.046(0.49)

0.115(0.92)

0.406(1.89)

0.826(2.23)

FFC-Adjusted Excess Return

-0.548(2.51)

-0.375(3.22)

-0.288(3.37)

-0.210(3.58)

-0.206(4.37)

-0.131(2.97)

-0.051(0.97)

0.084(0.89)

0.207(1.68)

0.499(2.29)

1.046(2.87)

Bond Illiquidity-Beta Sorted Portfolios, monthly returns, 24-month rolling window

H1: Small size funds with higher bond illiquidity beta portfolios have higher risk-adjusted returns compared to medium and large funds with high/low beta portfolios

H2: Funds with higher expenses and higher bond illiquidity beta portfolio have higher risk adjusted returns compared to funds with low expenses and low/high beta portfolios

H3: Funds with higher turnover and higher bond illiquidity beta portfolio have higher risk adjusted returns compared to funds with low turnover and low/high beta portfolios

Active Equity Mutual Funds and Funding Liquidity: Hypotheses


Fund Size and Bond Illiquidity Risk Portfolios



Size 1[Small 2 3 4

Size 5[Big]

Small-Big Size 1 2 3 4 Size 5

Small-Big

Beta 1 [low] -0.429 -0.567 -0.524 -0.443 -0.313 0.116 Beta 1 2.46 3.14 2.95 2.51 2.21 1.72

2 -0.261 -0.296 -0.310 -0.252 -0.171 0.90 2 3.51 3.73 3.61 3.39 2.49 1.51

3 -0.182 -0.213 -0.182 -0.170 -0.117 0.065 3 2.85 4.14 3.78 3.64 2.63 1.03

4 0.001 0.070 -0.034 0.017 0.056 0.054 4 0.02 0.57 0.48 0.24 1.03 0.84

Beta 5 [high] 0.432 0.384 0.269 0.313 0.379 -0.053 Beta 5 2.38 2.06 1.59 1.78 2.73 0.68

High-Low 0.861 0.951 0.793 0.756 0.692

High-Low 2.81 3.13 2.82 2.63 2.76

Fund Expenses and Bond Illiquidity Risk Portfolios



Expense1

[Small 2 3 4

Expense5

[Big]Small-

Big Expen

se1 2 3 4Expen

se5Small-

Big

Beta 1 [low] -0.341 -0.438 -0.524 -0.503 -0.473 -0.132 Beta 1 2.94 2.81 3.31 2.92 2.05 0.79

2 -0.231 -0.304 -0.292 -0.217 -0.236 -0.005 2 3.47 4.03 3.96 2.49 2.12 0.05

3 -0.183 -0.137 -0.192 -0.122 -0.147 0.037 3 3.67 2.94 3.72 1.68 2.40 0.43

4 -0.053 -0.007 0.016 0.070 0.034 0.087 4 1.03 0.13 0.23 0.81 0.31 0.74

Beta 5 [high] 0.227 0.247 0.370 0.365 0.524 0.297 Beta 5 1.80 1.70 2.18 1.99 2.19 2.00

High-Low 0.567 0.685 0.894 0.869 0.997

High-Low 2.57 2.56 3.19 2.98 2.75

Fund Turnover and Bond Illiquidity Risk Portfolios



Turnover1

[Small] 2 3 4

Turnover5

[Big]Small-

Big Turnover1 2 3 4

Turnover5

Small-Big

Beta 1 [low] -0.469 -0.503 -0.383 -0.425 -0.426 0.043 Beta 1 3.54 3.73 2.81 2.27 1.74 0.20

2 -0.322 0.313 -0.267 -0.188 -0.127 0.195 2 3.35 4.16 3.85 2.31 0.85 1.06

3 -0.234 0.231 -0.234 -0.119 0.048 0.282 3 4.28 4.72 4.78 2.10 0.37 1.76

4 -0.094 0.091 -0.061 0.107 0.245 0.340 4 1.55 1.79 0.95 0.82 1.76 2.07

Beta 5 [high] 0.199 0.160 0.242 0.359 0.546 0.347 Beta 5 1.72 1.15 1.63 1.83 2.54 2.40

High-Low 0.667 0.663 0.625 0.785 0.971

High-Low 2.91 2.69 2.52 2.65 2.87

The effect of Bond Illiquidity Risk on fund performance: Panel Regression


Alpha Alpha

Bond ILLIQ betat-1 0.635(2.07)

0.659(1.97)

MKT betat-1 -2.101(0.82)

SMB betat-1 3.128(0.92)

HML betat-1 -1.173(0.30)

UMD betat-1 -16.448(1.05)

Expensest-1 -0.705(1.46)

-0.520(1.19)

Log(TNA)t-1 -0.936(2.17)

-0.688(1.38)

Log(TNA)2t-1 0.058

(1.69)0.051(1.40)

Turnovert-1 -0.002(0.31)

0.007(1.79)

Log(Fund Age)t-1 -0.086(0.38)

0.065(0.24)

Log(Manager Tenure)t-1

0.271(1.44)

0.108(0.74)

Alphat-1 0.561(6.51)

0.529(3.44)

R2 0.30 0.31

This paper links Treasury bond illiquidity and funding liquidity

Applications: stocks with higher bond illiquidity betas earn higher risk-adjusted returns compared to lower bond illiquidity beta stocks (49 basis points per month)

Bond illiquidity beta has positive and significant risk premium after controlling for Fama-French and Carhart factors and firm characteristics such as size, book-to-market and stock illiquidity.

Funds in the highest bond illiquidity beta decile outperform funds in the lowest decile by about 80 basis points per month.

Smaller funds and higher bond illiq-beta – higher alpha

Higher expenses and higher bond illiq-beta – higher alpha

Higher turnover and higher bond illiq-beta – higher alpha

Conclusions


Ruslan Goyenko

Documents

Transcript of Ruslan Goyenko