Are Prices in OTC Transactions Different from Exchanges? The Case of - - PowerPoint PPT Presentation

▶
are prices in otc transactions different from exchanges
SMART_READER_LITE
LIVE PREVIEW

Are Prices in OTC Transactions Different from Exchanges? The Case of - - PowerPoint PPT Presentation

Oct 27, 2022 461 likes •678 views

Motivation Data Price Comparison Determinants Conclusion Are Prices in OTC Transactions Different from Exchanges? The Case of the Bund de Roure, Moench, Pelizzon and Schneider Frankfurt School of Finance, Bundesbank, Goethe University,

slide-1
SLIDE 1

Motivation Data Price Comparison Determinants Conclusion

Are Prices in OTC Transactions Different from Exchanges? The Case of the Bund

de Roure, Moench, Pelizzon and Schneider

Frankfurt School of Finance, Bundesbank, Goethe University, Scuola Normale Pisa c.deoure@fs.de

August 10, 2016

slide-2
SLIDE 2

Motivation Data Price Comparison Determinants Conclusion

Motivation

German Bund is the reference market in the Euro area and considered its most liquid market; This conclusion is mainly based on futures market. What about its underlying? The cash market is only observed through interdealer platforms (as MTS), where price discrimination does not

  • ccur;

However, the Bund cash market is mainly OTC; How does the Bund OTC market compare to exchange? Are trades in interdealer platforms comparable to OTC? Note: I use the term Bund for Bunds (10 and 30 years) and Bolbs (5 years).

slide-3
SLIDE 3

Motivation Data Price Comparison Determinants Conclusion

Average OTC price versus MTS

  • 97

98 99 100 101 102 103 Jan 01 Jan 15 Feb 01 Feb 15 Mar 01 Mar 15 Apr 01

Price (EUR per 100EUR face value)

DE0001135341

Candle: min best bid, mean of best bid, mean of best ask, max best ask quote

slide-4
SLIDE 4

Motivation Data Price Comparison Determinants Conclusion

Motivation: The Bund Cash Market

However, the banks in our sample traded Bunds 55,554 times OTC in 2008 26% (14,601) of these transactions were realized outside the bid-ask band in MTS

slide-5
SLIDE 5

Motivation Data Price Comparison Determinants Conclusion

Motivation: The Bund Cash Market

Large fragmentation (virtually every European bank hold Bunds), decentralized (37 dealers in the primary market) and well-established repo market increase costs of taking and reversing positions (Bundesbank 2007, Upper and Werner 2007) Bulk of trading and price discovery takes place in the futures market Daily turnover cash market 19 billion e vs. 70 billion e in the futures market (Finanzagentur)

slide-6
SLIDE 6

Motivation Data Price Comparison Determinants Conclusion

Data

MTS (about 2% of total market based on turnover) Largest interdealer platform Few trades but very active limit order book of standing executable quotes Data: best-bid /best-ask, turnover BaFin (about 15% of total market based on turnover) All fixed income trades of 27 banks between Jan, 1st and Dec, 31st 2008; Focus on OTC trades in own account Every trade is reported once by the initiator Data: price, volume, initiator, counterparty, trade sign (buy/sell)

slide-7
SLIDE 7

Motivation Data Price Comparison Determinants Conclusion

Summary Statistics

pre-Lehman post-Lehman 5y 10y 30y 5y 10y 30y #bonds 8 20 9 8 20 9 OTC #trades 11,641 32,401 2,882 3,153 6,697 1,064 avg volume (million e) 8.94 8.03 6.95 5.10 6.70 6.31 monthly turnover (million e) 11,200 28,200 1,720 7,300 19,200 3,360 MTS #trades 2,749 6,414 1,793 308 649 490 avg volume (million e) 7.74 6.80 3.80 8.54 7.58 3.95 monthly turnover (million e) 295 254 106 80 65 80 avg bid-ask spread (e) 0.043 0.045 0.388 0.200 0.289 0.949 avg quote size (million e) 15.6 15.1 4.4 10.3 9.8 4.2

slide-8
SLIDE 8

Motivation Data Price Comparison Determinants Conclusion

Are OTC trades outside the bid-ask band from MTS?

Comparing to best bid dbid = bestbid − pOTC Comparing to best ask dask = pOTC − bestask If dbid > 0 or dask > 0 trade is executed outside bid-ask band.

slide-9
SLIDE 9

Motivation Data Price Comparison Determinants Conclusion

Are OTC trades outside the bid-ask band from MTS?

Full Sample

5 10 15 Density

  • .6
  • .4
  • .2

.2 d_ask

(a) dask = pOTC − bestask

5 10 15 Density

  • .8
  • .6
  • .4
  • .2

.2 d_bid

(b) dbid = bestbid − pOTC

slide-10
SLIDE 10

Motivation Data Price Comparison Determinants Conclusion

Are OTC trades outside the bid-ask band from MTS?

Comparable size, before Lehman

5 10 15 Density

  • .6
  • .4
  • .2

.2 d_ask

(a) dask = pOTC − askbest

5 10 15 Density

  • .8
  • .6
  • .4
  • .2

.2 d_bid

(b) dbid = bidbest − pOTC

Comparable size, after Lehman

1 2 3 4 5 Density

  • .6
  • .4
  • .2

.2 d_ask

(c) dask = pOTC − askbest

2 4 6 Density

  • .8
  • .6
  • .4
  • .2

.2 d_bid

(d) dbid = bidbest − pOTC

slide-11
SLIDE 11

Motivation Data Price Comparison Determinants Conclusion

What drives trades outside the bid-ask band?

However, for every trade outside the bid-ask band there is someone doing a good and someone a bad job. What drives trades outside the bid-ask band?

slide-12
SLIDE 12

Motivation Data Price Comparison Determinants Conclusion

Define Good/Bad Buys, Good/Bad Sells

MTS best bid MTS best ask price discount premium

Good Buys Bad Sells Bad Buys Good Sells

slide-13
SLIDE 13

Motivation Data Price Comparison Determinants Conclusion

Summary Statistics

pre-Lehman post-Lehman 5y 10y 30y 5y 10y 30y trades inside spread 67.79% 64.14% 75.18% 87.44% 86.90% 89.66% Good Buys 7.83% 10.30% 9.30% 3.55% 4.88% 4.61% Bad Sells 12.44% 11.73% 6.18% 5.93% 5.44% 2.44% Bad Buys 5.94% 6.60% 3.30% 1.49% 2.57% 1.41% Good Sells 6.00% 7.23% 6.04% 1.59% 3.21% 1.88%

slide-14
SLIDE 14

Motivation Data Price Comparison Determinants Conclusion

Estimation

Extensive margin: what is the probability that a bank does a good/bad trade? (our focus today) Instensive margin: what are the drivers of the OTC premium?

slide-15
SLIDE 15

Motivation Data Price Comparison Determinants Conclusion

Probit Good Buy, below best-bid (Marg. Effect at Means)

Volume 0.004*** 0.002*** 0.002** 0.001 0.003 (0.001) (0.001) (0.001) (0.001) (0.003) Bid-Ask Spread

  • 0.529***
  • 0.491***
  • 0.645***
  • 0.504***

(0.032) (0.063) (0.124) (0.080) After-Lehman

  • 0.011***

(0.001) Volume*After-Lehman 0.005*** (0.001) Bid-Ask Spread*After-Lehman 0.003*** (0.010) Dealer 0.013 (0.052) Volume*Dealer

  • 0.002

(0.003) Bid-Ask Spread*Dealer 0.024 (0.051) Interdealer

  • 0.357***

(0.098) Volume*Interdealer 0.023*** (0.007) Bid-Ask Spread*Interdealer

  • 0.757

(0.524) Security FE yes yes yes yes yes Bank FE no no yes yes no Time FE no no weekly no weekly Pseudo R2 0.0109 0.0456 0.0574 0.0551 0.0517 # Obs 55552 55552 55541 55541 55552

slide-16
SLIDE 16

Motivation Data Price Comparison Determinants Conclusion

Probit Bad Sell, below best-bid (Marg. Effect at Means)

Volume 0.012*** 0.011*** 0.008** 0.008** 0.003 (0.001) (0.001) (0.003) (0.004) (0.001) Bid-Ask Spread

  • 0.329***
  • 0.266***
  • 0.387***
  • 0.293***

(0.083) (0.025) (0.035) (0.023) After-Lehman

  • 0.028

(0.024) Volume*After-Lehman

  • 0.001

(0.002) Bid-Ask Spread*After-Lehman 0.177*** (0.024) Dealer

  • 0.266***

(0.039) Volume*Dealer 0.014*** (0.002) Bid-Ask Spread*Dealer 0.065 (0.055) Interdealer 0.174** (0.074) Volume*Interdealer

  • 0.009*

(0.005) Bid-Ask Spread*Interdealer

  • 0.120

(0.112) Security FE yes yes yes yes yes Bank FE no no yes yes no Time FE no no weekly no weekly Pseudo R2 0.0436 0.0694 0.1367 0.1317 0.1156 # Obs 55477 55477 55477 55477 55477

slide-17
SLIDE 17

Motivation Data Price Comparison Determinants Conclusion

Probit Bad Buy, above best-ask (Marg. Effect at Means)

Volume 0.013*** 0.012*** 0.009*** 0.010*** 0.005*** (0.001) (0.001) (0.002) (0.002) (0.002) Bid-Ask Spread

  • 0.301***
  • 0.293***
  • 0.347***
  • 0.265***

(0.042) (0.005) (0.062) (0.039) After-Lehman

  • 0.022

(0.022) Volume*After-Lehman

  • 0.002

(0.001) Bid-Ask Spread*After-Lehman 0.101*** (0.035) Dealer

  • 0.162***

(0.038) Volume*Dealer 0.008*** (0.002) Bid-Ask Spread*Dealer

  • 0.085

(0.063) Interdealer 0.457*** (0.096) Volume*Interdealer

  • 0.027***

(0.006) Bid-Ask Spread*Interdealer

  • 0.427**

(0.216) Security FE yes yes yes yes yes Bank FE no no yes yes no Time FE no no weekly no weekly Pseudo R2 0.0565 0.0788 0.1388 0.1325 0.1176 # Obs 55452 55452 55365 55365 55452

slide-18
SLIDE 18

Motivation Data Price Comparison Determinants Conclusion

Probit Good Sell, above best-ask (Marg. Effect at Means)

Volume

  • 0.002*
  • 0.004***
  • 0.003**
  • 0.000
  • 0.002

(0.001) (0.001) (0.001) (0.002) (0.003) Bid-Ask Spread

  • 0.551***
  • 0.666***
  • 0.787***
  • 0.78***

(0.028) (0.094) (0.168) (0.117) After-Lehman 0.154*** (0.016) Volume*After-Lehman

  • 0.011***

(0.001) Bid-Ask Spread*After-Lehman 0.158 (0.115) Dealer 0.007 (0.056) Volume*Dealer

  • 0.001

(0.003) Bid-Ask Spread*Dealer 0.142*** (0.047) Interdealer

  • 0.396*

(0.182) Volume*Interdealer 0.022* (0.012) Bid-Ask Spread*Interdealer 0.356* (0.193) Security FE yes yes yes yes yes Bank FE no no yes yes no Time FE no no weekly no weekly Pseudo R2 0.0109 0.0414 0.0546 0.0508 0.0486 # Obs 55552 55552 55541 55541 55541

slide-19
SLIDE 19

Motivation Data Price Comparison Determinants Conclusion

Conclusion

OTC prices can differ substantially from interdealer platform Trade size and market liquidity drive trades outside the bid-ask band We find evidence for price differentiation:

1

Interdealer trade is less likely to be good and more likely to be bad

2

Dealer/Non-dealer trade: less likely to be bad

Lehman increases bid-ask spread and reduces the number of trades outside bid-ask band

slide-20
SLIDE 20

Motivation Data Price Comparison Determinants Conclusion

Next Steps

Change year of the analysis to avoid bias Include full LOB instead of best bid and best ask Second precision Clear cut research question: Why do banks trade OTC in a liquid market?