Popular in Economics

Central Monetary Authorities

Macroeconomic modelling of CBDC: a critical review

European Central Bank Ulrich Bindseil

Central Monetary Authorities

Financing Private Credit

Federal Reserve Bank of New York Nina Boyarchenko

Economics

Failing Banks

Massachusetts Institute of Technology Emil Verner

Economic History

Golden Fetters

UC Berkeley Barry Eichengreen

Economic Inequalities

Visions of Inequality

City University of New York Branko Milanovic

Economics

School Finance Equalization Increases Intergenerational Mobility

Yale University Barbara Biasi

Economics

Olivier Blanchard on U.S. Pandemic-Era Inflation

Peterson Institute for International Economics Olivier Blanchard

The Effect of Primary Dealer Constraints on Intermediation in the Treasury Market

View more insights in Social Sciences - Economics - Central Monetary Authorities - Macroeconomics

View more figures from Federal Reserve Bank of Boston

View the page of Falk Bräuning

View the page of Hillary Stein

Read the Study

Next in Central Monetary Authorities

xtevent: Estimation and Visualization in the Linear Panel Event-Study Design

Simon Freyaldenhoven | Federal Reserve Bank of Philadelphia

Time-varying Persistence of House Price Growth

C.Y. Choi | Federal Reserve Bank of Dallas

Falk Bräuning and Hillary Stein discuss the extent to which the Treasury securities market can be affected by regulatory constraints on primary dealers, a select group of market-making financial firms that are typically affiliated with large banking organizations.

The views expressed in this video are solely those of the author(s) and do not represent the views of the Federal Reserve Bank of Boston, the principals of the Board of Governors, or the Federal Reserve System.

Image courtesy of interviewee. October 22, 2024

Log-in or Sign-up to Faculti
Currently viewing this subject insight as a guest. You have insight(s) remaining for this month. Login to view 8000+ figures on the platform.

Generating Workspace...

Export as PDF

This video discusses the significant role that U.S. Treasury securities play in the global economy, primarily as a means of financing the U.S. government and as a benchmark for risk-free investments. It explores the impact of constraints faced by primary dealers—financial firms that facilitate trading in Treasury securities—on the market's liquidity, turnover, and pricing.

When these dealers experience tighter constraints, they tend to reduce their Treasury holdings, leading to decreased liquidity and increased spreads. The research highlights the importance of understanding the constraints on these intermediaries, particularly in light of findings from the 2008 financial crisis and subsequent regulatory changes.

The study utilizes detailed data from two Federal Reserve datasets to analyze how constraints affect primary dealers' trading behavior. It distinguishes between two types of constraints: the Supplementary Leverage Ratio (SLR) and internal Value at Risk (VaR) limits. The findings suggest that easing constraints, such as exempting Treasury holdings from SLR calculations, results in primary dealers increasing their Treasury positions and turnover while lowering spreads, indicating improved market liquidity. In contrast, tightening VaR limits leads to reduced positions and reduced liquidity.

Overall, the analysis emphasizes the criticality of primary dealers and their regulatory constraints on the liquidity and functioning of Treasury markets, especially in an environment of rising government debt. These insights are relevant for shaping monetary policy and financial regulations.

Language:

Export as PDF

Export as PDF
Here are five key points from the video:

1. **Role of Treasuries**: U.S. Treasuries are essential for financing the federal government, serve as significant investment instruments for global investors, act as risk-free benchmarks for financial institutions, and are critical for the Federal Reserve's implementation of monetary policy.

2. **Impact of Primary Dealers**: The Treasury market is predominantly intermediated by a select group of financial firms known as primary dealers. When these primary dealers face constraints (e.g., regulatory or internal risk limits), it can adversely affect their holdings of Treasuries and overall market liquidity.

3. **Research Methods and Data**: The research conducted utilizes two key datasets: the FR 2004, which records the positions and turnover of primary dealers, and the Regulation VV (VP1) dataset, which provides internal risk limits for trading desks. Empirical methodologies are utilized to assess how constraints impact treasury positions and liquidity.

4. **Findings on Constraints**: The study reveals that when primary dealers experience tighter constraints (like changes in Supplementary Leverage Ratios), they reduce their Treasury holdings and turnover, leading to lower market liquidity. Conversely, lifting these constraints (as seen during the 2020 SLR change) correlates with increased Treasury holdings and improved market liquidity.

5. **Heterogeneous Effects of Risk Constraints**: The research indicates that different types of securities respond variably to risk constraints. For instance, while value-at-risk limits impact holdings and turnover more significantly for higher volatility securities, the exemption of Treasury securities from leverage exposure broadly benefits all types of Treasuries, enhancing overall market liquidity.

Export as PDF
Here are the key quotes extracted from the video about the role of treasuries in the global economy and the challenges faced by primary dealers:

1. **Importance of Treasuries**: "The idea that treasuries play a crucial role in the global economy. They're the primary means of financing the US Federal Government."

2. **Investment Instrument**: "They're a significant investment instrument and hedging vehicle for global investors."

3. **Risk-free Benchmark**: "They're a risk-free benchmark for many other financial institutions."

4. **Monetary Policy Tool**: "They're an important tool for the Federal Reserves implementation of monetary policy."

5. **Role of Primary Dealers**: "For how important they are, they're intermediated by just a select group of market making financial firms, which are called primary dealers."

6. **Impact of Constraints**: "What happens when these primary dealers face constraints? How does that impact the treasury market?"

7. **Reduction in Treasury Holdings**: "When the primary dealers are faced with tighter constraints... these primary dealers do indeed reduce their treasury holdings."

8. **Lower Liquidity**: "This is accompanied by lower liquidity in the treasury market."

9. **Literature on Financial Intermediaries**: "Our work shows that... financial intermediaries' constraints really are important for driving asset prices."

10. **Unique Data Contribution**: "We have a very unique, very detailed, granular data on individual actors... to disentangle how those constraints that these primary dealers face really drive their decisions."

11. **Types of Constraints**: "You can imagine constraints on the overall bank balance sheet... or especially relevant for these primary dealers on the actual trading desk themselves."

12. **Impact of the SLR Constraint**: "SLR constrained banks might be constrained in their treasury positions."

13. **Findings Post-COVID Policy Change**: "If this constraint was lifted... primary dealers... increase their position in treasuries after the policy change."

14. **Increased Turnover and Lower Spreads**: "We also see that their margin... declines... primary dealer associated with a bank leverage ratio, a one percentage point lower leverage ratio... see an increase in about nine percentage points."

15. **Market Liquidity Effects**: "Once you lift those binding constraints... the market becomes more liquid in some sense."

16. **Risk-Sensitive Value at Risk Constraints**: "As value at risk limits become tighter, the primary dealers reduce their positions in treasury securities."

17. **Importance of Treasury Markets**: "Treasury markets are at the heart of the financial system, both in the United States and globally."

18. **Regulatory Perspective**: "It's important to understand the impact of regulatory constraints... ultimately, this intermediation capacity we show is crucial for treasury market liquidity more broadly."

These quotes encapsulate the significance of treasuries, the operational mechanisms of primary dealers, the impacts of regulatory constraints, and the overall implications for market liquidity.

X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:90000

1
00:00:00.145 --> 00:00:02.325
The idea that treasuries play a crucial

2
00:00:02.355 --> 00:00:03.525
role in the global economy.

3
00:00:03.755 --> 00:00:04.965
They're the primary means

4
00:00:04.965 --> 00:00:06.885
of financing the US Federal Government.

5
00:00:07.265 --> 00:00:10.045
Um, they're a significant investment instrument in hedging

6
00:00:10.045 --> 00:00:11.245
vehicle for global investors.

7
00:00:11.245 --> 00:00:12.525
They're risk-free benchmark

8
00:00:12.545 --> 00:00:14.125
for many other financial institutions,

9
00:00:14.125 --> 00:00:16.005
and they're an important tool for the Federal Reserves

10
00:00:16.005 --> 00:00:17.485
implementation of monetary policy.

11
00:00:18.185 --> 00:00:21.485
Um, that said, for how important they are,

12
00:00:21.555 --> 00:00:23.885
they're intermediated by just a select group

13
00:00:23.885 --> 00:00:25.965
of market making financial firms,

14
00:00:25.975 --> 00:00:27.685
which are called primary dealers.

15
00:00:28.025 --> 00:00:30.205
Um, and these primary dealers are typically affiliated

16
00:00:30.205 --> 00:00:31.805
with large banking organizations.

17
00:00:32.185 --> 00:00:34.325
So the question was, and the idea was,

18
00:00:34.635 --> 00:00:39.365
what happens when these primary dealers face,

19
00:00:39.905 --> 00:00:41.685
constraints?

20
00:00:41.865 --> 00:00:44.085
How does that impact the treasury market?

21
00:00:44.185 --> 00:00:46.085
And our work shows that when,

22
00:00:46.085 --> 00:00:49.445
the primary dealers are faced with tighter constraints,

23
00:00:49.445 --> 00:00:51.045
and we can sort of talk about,

24
00:00:51.645 --> 00:00:52.765
defining what we mean by constraints.

25
00:00:53.105 --> 00:00:56.005
Um, these primary dealers do indeed reduce their treasury

26
00:00:56.045 --> 00:00:57.925
holdings, and this is accompanied

27
00:00:57.925 --> 00:01:00.045
by lower liquidity in the treasury market.

28
00:01:00.365 --> 00:01:02.725
Turnover decreases and bees price increase. Our

29
00:01:02.725 --> 00:01:04.725
Work really, is, is,

30
00:01:04.785 --> 00:01:07.325
is grounded in this relatively large literature,

31
00:01:07.325 --> 00:01:10.045
especially theoretical literature on the role

32
00:01:10.045 --> 00:01:12.725
of financial intermediaries, in asset pricing

33
00:01:13.385 --> 00:01:16.605
and specifically, there's a relatively large literature

34
00:01:16.605 --> 00:01:20.125
that focuses on what constraint to these constraints

35
00:01:20.145 --> 00:01:22.485
to these intermediaries really do, in terms

36
00:01:22.485 --> 00:01:24.205
of asset pricing and liquidity.

37
00:01:24.945 --> 00:01:28.805
So part of the literature, was really inspired

38
00:01:28.825 --> 00:01:30.605
by the events of the great financial crisis.

39
00:01:30.705 --> 00:01:32.805
So we've seen a surge in, in these,

40
00:01:33.825 --> 00:01:36.365
mostly theoretical papers in the aftermath.

41
00:01:36.585 --> 00:01:37.845
Um, therefore,

42
00:01:38.025 --> 00:01:41.205
but, there's also a host of papers that tries

43
00:01:41.305 --> 00:01:43.605
to get at these questions empirically.

44
00:01:43.825 --> 00:01:46.885
I'm trying to find empirical support for the theories that,

45
00:01:46.885 --> 00:01:49.605
that argue that financial intermediaries constraints really

46
00:01:49.605 --> 00:01:51.765
are important for driving asset prices.

47
00:01:52.555 --> 00:01:55.245
What we add relative to the existing,

48
00:01:55.525 --> 00:01:58.165
literature is really that we have a very unique,

49
00:01:58.275 --> 00:02:01.685
very detailed, granular data on individual actors

50
00:02:01.755 --> 00:02:03.885
constraints in the, in the treasury market.

51
00:02:04.025 --> 00:02:05.445
So we're really able

52
00:02:05.505 --> 00:02:07.885
to go really into the nitty gritty details

53
00:02:07.985 --> 00:02:12.205
and really disentangle how those, constraints

54
00:02:12.205 --> 00:02:14.525
that these primary dealer face really,

55
00:02:14.815 --> 00:02:18.645
drive their decisions in terms of, treasury holdings,

56
00:02:18.645 --> 00:02:20.725
in terms of, turnover, in terms of,

57
00:02:20.825 --> 00:02:21.925
spread they charge.

58
00:02:22.385 --> 00:02:24.565
So we are really, adding to this literature

59
00:02:24.565 --> 00:02:28.245
by providing these very detailed micro data that allows us

60
00:02:28.245 --> 00:02:31.045
to costly identify, these effects.

61
00:02:31.625 --> 00:02:35.525
We use, two, supervisory,

62
00:02:35.705 --> 00:02:38.405
or two sort of confidential, federal Reserve,

63
00:02:38.405 --> 00:02:40.525
data sets, one of which is, supervisory.

64
00:02:40.705 --> 00:02:44.925
We use, the FR 2004,

65
00:02:45.745 --> 00:02:48.285
which is the government securities dealer reports.

66
00:02:48.545 --> 00:02:51.805
And we use the, regulation VV quantitative,

67
00:02:51.805 --> 00:02:55.205
measurements, which we call the, the VP one, dataset.

68
00:02:55.265 --> 00:02:59.565
So the, FR 2004, that's a, sort

69
00:02:59.565 --> 00:03:01.885
of longer historical data set,

70
00:03:01.885 --> 00:03:04.805
that includes information on the outright positions

71
00:03:04.805 --> 00:03:07.165
and the turnover of all primary dealers

72
00:03:07.165 --> 00:03:08.485
of US government securities.

73
00:03:08.715 --> 00:03:12.645
This data set includes, the important response variables

74
00:03:12.645 --> 00:03:16.765
that I talked about earlier, of, positions

75
00:03:16.865 --> 00:03:19.645
and turnovers of these primary dealers.

76
00:03:19.745 --> 00:03:21.765
The second data set that we use it,

77
00:03:21.765 --> 00:03:25.685
the FFR VV one provides desk specific internal,

78
00:03:26.595 --> 00:03:29.525
risk limits for the trading desks

79
00:03:29.625 --> 00:03:31.605
of large banking organizations.

80
00:03:32.105 --> 00:03:34.445
So, this data set is relatively newer.

81
00:03:34.555 --> 00:03:39.005
It's been collected since July, 2014,

82
00:03:39.425 --> 00:03:44.285
as part of the 2013, VUCA rule of the Dodd-Frank Act.

83
00:03:44.505 --> 00:03:46.645
The VUCA rule prohibits,

84
00:03:46.815 --> 00:03:49.045
banks from engaging in proprietary trading,

85
00:03:49.105 --> 00:03:51.165
but as part of that rule, banks are required

86
00:03:51.165 --> 00:03:54.365
to report their internal organization of trading desks,

87
00:03:54.555 --> 00:03:55.885
desk names, descriptions,

88
00:03:55.945 --> 00:04:00.205
and also report the, internal risk limits,

89
00:04:00.665 --> 00:04:02.405
of these trading desks.

90
00:04:02.505 --> 00:04:04.805
In practice. These risk limits take a couple forms.

91
00:04:05.025 --> 00:04:06.685
Um, they can be sensitivity of limits,

92
00:04:06.715 --> 00:04:08.565
they can be value at risk or VAR limits.

93
00:04:08.625 --> 00:04:11.085
And we primarily focus on the VAR limits.

94
00:04:11.425 --> 00:04:14.605
Um, we compliment these two sort of main micro data sets

95
00:04:14.795 --> 00:04:16.725
with our Y nine C,

96
00:04:17.105 --> 00:04:18.965
and then, auction data as well

97
00:04:18.965 --> 00:04:20.005
as, some pricing data.

98
00:04:20.265 --> 00:04:22.765
The question is, what happens,

99
00:04:23.675 --> 00:04:26.725
when banks face tighter constraints?

100
00:04:27.025 --> 00:04:28.565
Um, what happens to their positions?

101
00:04:28.565 --> 00:04:30.485
What happens to turnover? What happens to the best spreads

102
00:04:30.485 --> 00:04:32.005
and treasury market liquidity?

103
00:04:32.465 --> 00:04:37.205
Um, so, the, I think let's, we can back up

104
00:04:37.205 --> 00:04:39.005
for a second and ask the question of

105
00:04:39.005 --> 00:04:40.045
what do we mean by constraints?

106
00:04:40.545 --> 00:04:44.965
Um, so banks, have sort of a variety.

107
00:04:44.995 --> 00:04:47.245
They have these, this organizational structure,

108
00:04:47.245 --> 00:04:49.125
which is composed of different organizational layers,

109
00:04:49.125 --> 00:04:51.325
and you can imagine constraints at different

110
00:04:51.325 --> 00:04:52.965
points of the banking structure.

111
00:04:53.025 --> 00:04:55.045
You can imagine constraints on the,

112
00:04:55.275 --> 00:04:56.645
overall bank balance sheet

113
00:04:57.105 --> 00:05:00.485
or, especially relevant for these,

114
00:05:00.485 --> 00:05:02.325
primary dealers which are trading these,

115
00:05:02.325 --> 00:05:03.805
the government securities,

116
00:05:04.065 --> 00:05:05.565
on the actual trading desk themselves.

117
00:05:06.025 --> 00:05:10.645
Um, and so because banks can face constraints in different

118
00:05:10.645 --> 00:05:14.525
places, we actually use two different methods to

119
00:05:15.195 --> 00:05:17.005
explore, the impact

120
00:05:17.185 --> 00:05:18.645
of two different types of constraints.

121
00:05:19.185 --> 00:05:23.965
Um, so first we start with the SLR constraint, which,

122
00:05:24.185 --> 00:05:26.925
is part of the US implementation

123
00:05:26.925 --> 00:05:28.485
of the Basel three accords.

124
00:05:28.825 --> 00:05:31.965
Um, and the idea here is, the SLR is the ratio

125
00:05:31.965 --> 00:05:34.285
of tier one capital to total leverage exposure,

126
00:05:34.515 --> 00:05:39.325
because exposure includes gross positions in securities,

127
00:05:39.595 --> 00:05:40.965
held by primary dealers.

128
00:05:41.065 --> 00:05:43.845
Um, SLR constrained banks might be constrained in

129
00:05:43.845 --> 00:05:45.365
their treasury positions.

130
00:05:45.425 --> 00:05:47.845
Um, so that's sort of, that's constraint one,

131
00:05:47.865 --> 00:05:51.365
and we're gonna use a particular method to,

132
00:05:51.625 --> 00:05:53.565
to explore the impact of constraint one,

133
00:05:53.565 --> 00:05:56.565
constraint two is these internal constraints on banks

134
00:05:56.565 --> 00:05:57.645
trading activities.

135
00:05:58.265 --> 00:06:02.485
Um, and so as I mentioned, the ULCA rule,

136
00:06:03.155 --> 00:06:05.205
does limit banks from taking too much risk

137
00:06:05.205 --> 00:06:06.725
and engaging in proprietary trading.

138
00:06:06.745 --> 00:06:09.525
But separate from that, even though treasuries aren't,

139
00:06:10.265 --> 00:06:14.725
they're exempt from the, from the ULCA rule,

140
00:06:15.295 --> 00:06:18.525
banks still have internal risk limits

141
00:06:18.585 --> 00:06:20.965
and they're still not gonna let their trading desks take,

142
00:06:21.265 --> 00:06:24.325
you know, unlimited risk and, and treasuries.

143
00:06:24.325 --> 00:06:27.885
And we actually can see in our data that there are,

144
00:06:27.885 --> 00:06:29.645
value at risk constraints on the trading desk

145
00:06:29.645 --> 00:06:31.885
that only trade treasuries.

146
00:06:31.985 --> 00:06:34.805
And so we have a second method for looking at the impact

147
00:06:34.945 --> 00:06:36.805
of these, internal,

148
00:06:37.485 --> 00:06:39.685
VR risk limits on the, on the training desk.

149
00:06:39.835 --> 00:06:43.615
When we look at the SLR, the impact of SLR constraints,

150
00:06:44.195 --> 00:06:47.415
we use a difference in difference methodology,

151
00:06:47.545 --> 00:06:52.015
which leverages the fact that in April, 2020,

152
00:06:52.545 --> 00:06:56.855
there was a surprise, sort of relaxation

153
00:06:57.235 --> 00:07:00.295
of this SLR constraint,

154
00:07:01.985 --> 00:07:05.285
in which treasuries were exempt from the SLR.

155
00:07:05.665 --> 00:07:08.725
Um, and so what we do is we compare, banks

156
00:07:08.725 --> 00:07:09.805
that were more constrained

157
00:07:09.805 --> 00:07:14.165
or less constrained in their SLR according to the 2019,

158
00:07:14.665 --> 00:07:17.045
Q4 value of their SLR.

159
00:07:17.305 --> 00:07:21.445
We compare, how more constrained banks reacted

160
00:07:21.455 --> 00:07:22.525
after this,

161
00:07:23.045 --> 00:07:25.245
surprise treasury exemption from the SLR versus

162
00:07:25.275 --> 00:07:26.445
less constrained banks.

163
00:07:26.865 --> 00:07:29.485
Um, so that's the first methodology,

164
00:07:29.865 --> 00:07:33.605
and, that allows us to look at sort

165
00:07:33.605 --> 00:07:35.765
of differences in positions between more constrained

166
00:07:35.765 --> 00:07:37.925
and less constrained banks when that is relaxed.

167
00:07:38.625 --> 00:07:43.285
Um, our second, method,

168
00:07:43.945 --> 00:07:46.725
uses the granular instrumental,

169
00:07:47.705 --> 00:07:48.845
or something sort of similar

170
00:07:48.865 --> 00:07:51.005
to the granular instrumental variable approach.

171
00:07:51.145 --> 00:07:54.045
And the idea here is that if only,

172
00:07:54.825 --> 00:07:57.565
one trading desk has a risk limit change,

173
00:07:58.115 --> 00:08:02.695
that risk limit change is, probably exogenous

174
00:08:02.755 --> 00:08:03.975
to overall market conditions.

175
00:08:04.235 --> 00:08:05.855
And so you can look at these sort of,

176
00:08:06.185 --> 00:08:07.335
these risk limit changes

177
00:08:07.445 --> 00:08:09.055
that are only happening at one desk,

178
00:08:09.155 --> 00:08:10.975
and you do that by running this,

179
00:08:11.125 --> 00:08:13.295
saturated fixed effects regression.

180
00:08:13.875 --> 00:08:17.215
Um, and then you aggregate all

181
00:08:17.215 --> 00:08:20.535
of these idiosyncratic changes up,

182
00:08:20.715 --> 00:08:24.495
and that gets this exogenous,

183
00:08:25.555 --> 00:08:28.855
risk limit shock.

184
00:08:29.115 --> 00:08:30.295
Um, and then we can sort

185
00:08:30.295 --> 00:08:33.975
of regress this exogenous risk limit shock,

186
00:08:34.395 --> 00:08:38.735
or look, look, regress some positions and, turnover

187
00:08:39.195 --> 00:08:42.735
and, bitas spreads on this exogenous risk limit shock

188
00:08:42.805 --> 00:08:47.375
Looking essentially as at, at two depths of analysis here,

189
00:08:47.405 --> 00:08:50.095
like related to the different constraints

190
00:08:50.195 --> 00:08:52.215
and related the different methods here.

191
00:08:52.755 --> 00:08:55.845
If we're, I'm gonna first talk about the findings for the

192
00:08:56.355 --> 00:08:58.605
supplementary laboratory ratio constraint.

193
00:08:59.065 --> 00:09:03.905
So here we really exploit this, this policy change,

194
00:09:03.965 --> 00:09:05.705
in the wake of the covid,

195
00:09:06.425 --> 00:09:11.105
pandemic in early 2020 when the Federal Reserve, made a

196
00:09:11.745 --> 00:09:15.465
temporary, change to the SLR computation,

197
00:09:15.605 --> 00:09:19.425
in the following sense that, treasury exposure and,

198
00:09:19.445 --> 00:09:23.145
and also reserves held at the central banks were, as

199
00:09:23.145 --> 00:09:26.985
of April 1st, 2020 20, allowed

200
00:09:26.985 --> 00:09:29.785
to be exempt from the total leverage exposure.

201
00:09:30.045 --> 00:09:31.105
So what does it mean?

202
00:09:31.725 --> 00:09:34.105
It means that if a primary dealer affiliated

203
00:09:34.105 --> 00:09:38.275
with a bank holding company, holds a treasury,

204
00:09:38.995 --> 00:09:41.795
position in its book, that treasury position will not add

205
00:09:41.795 --> 00:09:42.955
to the leverage exposure,

206
00:09:43.185 --> 00:09:45.635
therefore will not be costly in terms of capital.

207
00:09:46.095 --> 00:09:47.675
And that change really was,

208
00:09:47.675 --> 00:09:49.635
announced like on a very short notice

209
00:09:49.655 --> 00:09:51.475
and became effective immediately.

210
00:09:51.935 --> 00:09:56.155
So we're looking around this, tight window of a couple

211
00:09:56.175 --> 00:09:59.195
of, weeks around before and after that policy change,

212
00:09:59.295 --> 00:10:03.115
and look how different dealers, respond to

213
00:10:03.115 --> 00:10:05.475
that change in terms of, holdings and,

214
00:10:05.535 --> 00:10:07.115
and turnover and spread.

215
00:10:07.115 --> 00:10:10.715
And so what we really find is like, very intuitive if

216
00:10:11.725 --> 00:10:13.535
this, constraint was lifted

217
00:10:13.755 --> 00:10:16.935
by exempting the treasury securities from the total leverage

218
00:10:17.255 --> 00:10:20.815
exposures, primary dealers that are affiliated

219
00:10:20.815 --> 00:10:23.975
with a bank holding company that has a relatively

220
00:10:24.595 --> 00:10:25.815
low leverage ratio

221
00:10:25.815 --> 00:10:29.215
that is relatively more constrained than other, other,

222
00:10:29.625 --> 00:10:34.055
other bank holding companies, those primary dealers, they,

223
00:10:34.195 --> 00:10:37.615
increase their position in treasuries after the,

224
00:10:37.615 --> 00:10:38.855
after the policy change,

225
00:10:39.515 --> 00:10:41.895
and they increase also their total turnover.

226
00:10:41.955 --> 00:10:46.175
And we also see that their, margin, their kind

227
00:10:46.175 --> 00:10:48.375
of a measure of intermediation spread that

228
00:10:48.375 --> 00:10:50.215
that OM declines essentially.

229
00:10:51.035 --> 00:10:53.815
Um, I can also put some precise numbers on it.

230
00:10:53.815 --> 00:10:55.495
These effects are relatively sizable.

231
00:10:55.555 --> 00:10:57.775
So our estimates indicate that,

232
00:10:58.525 --> 00:11:00.575
primary dealer associated with a bank

233
00:11:00.575 --> 00:11:03.975
with a one percentage points, lower

234
00:11:04.895 --> 00:11:08.455
leverage ratio, a one percentage point lower leverage ratio,

235
00:11:08.455 --> 00:11:12.415
after the policy policy change, those primary dealers,

236
00:11:12.475 --> 00:11:16.055
see an increase in about nine percentage points relative

237
00:11:16.055 --> 00:11:19.655
to other dealers in their pro treasury position holdings.

238
00:11:19.795 --> 00:11:23.015
We also see a, higher turnover and a, and a

239
00:11:23.015 --> 00:11:25.255
and a lower, bit are spread here.

240
00:11:25.595 --> 00:11:28.095
So this is all, according to the theory

241
00:11:28.125 --> 00:11:29.455
that these constraints matters.

242
00:11:29.515 --> 00:11:32.015
And once you lift those binding constraint,

243
00:11:32.015 --> 00:11:35.495
then there's a higher turnover, higher positions holding

244
00:11:35.555 --> 00:11:37.775
and a, a lower intermediation spread.

245
00:11:37.835 --> 00:11:40.455
So the, the market becomes more liquid in some sense.

246
00:11:41.605 --> 00:11:44.215
It's what is very interesting, is

247
00:11:44.725 --> 00:11:47.895
that we also looked at heterogeneous effect across different

248
00:11:48.135 --> 00:11:49.135
treasury securities.

249
00:11:49.915 --> 00:11:52.055
And our findings here show

250
00:11:52.765 --> 00:11:55.445
that there's actually no heterogeneous effects.

251
00:11:56.845 --> 00:11:59.195
After the policy change was enacted,

252
00:11:59.295 --> 00:12:02.475
the dealers increased their position holdings across all

253
00:12:02.475 --> 00:12:04.875
types of treasury securities, t-bills,

254
00:12:04.875 --> 00:12:07.355
and also longer dated, securities.

255
00:12:07.365 --> 00:12:09.995
Given the, the, the regulatory framework.

256
00:12:09.995 --> 00:12:11.115
This is very intuitive

257
00:12:11.115 --> 00:12:13.515
because these supplementary leverage ratio,

258
00:12:13.815 --> 00:12:15.395
as Hillary mentioned, this is the,

259
00:12:15.395 --> 00:12:17.875
implementation in the US of the basal accords.

260
00:12:18.225 --> 00:12:20.435
This is meant to be risk insensitive.

261
00:12:21.015 --> 00:12:22.995
So we are finding exactly,

262
00:12:23.145 --> 00:12:25.515
this risk insensitive, effects.

263
00:12:25.625 --> 00:12:28.835
When we look at the, what, what the, temporary exemption

264
00:12:28.835 --> 00:12:31.235
of treasuries that to the, to the holdings,

265
00:12:31.895 --> 00:12:34.755
the primary dealers increase the holdings across the board

266
00:12:34.755 --> 00:12:36.275
of all treasury securities.

267
00:12:36.275 --> 00:12:39.275
Where we look at the, at the supplementary leverage ratio,

268
00:12:40.625 --> 00:12:45.005
if you look at the effects of value at risk, shifts,

269
00:12:45.025 --> 00:12:48.365
as Hillary explained the, the method already, we are finding

270
00:12:49.545 --> 00:12:51.715
qualitatively very similar results.

271
00:12:52.615 --> 00:12:56.225
What we find is that as value at risk limits become tighter,

272
00:12:57.915 --> 00:13:00.295
the, primary dealers reduce their

273
00:13:01.015 --> 00:13:02.535
positions in treasury securities.

274
00:13:03.315 --> 00:13:07.815
As a result, we also see, total turnover going down and,

275
00:13:07.915 --> 00:13:11.095
and spreads, increase intermediation spreads increase.

276
00:13:11.095 --> 00:13:14.895
So this is like qualitatively very similar to what we find,

277
00:13:14.955 --> 00:13:17.135
for the supplementary leverage ratio.

278
00:13:18.065 --> 00:13:20.495
There is, one important difference

279
00:13:20.495 --> 00:13:22.895
that i I wanna highlight here, which is

280
00:13:22.895 --> 00:13:25.655
that the conceptually, the value at risk constraint,

281
00:13:25.795 --> 00:13:27.255
as the name already suggests this,

282
00:13:27.535 --> 00:13:29.255
are risk sensitive constraints.

283
00:13:29.315 --> 00:13:31.415
So a security with

284
00:13:31.425 --> 00:13:34.655
where may add different value at risk,

285
00:13:34.655 --> 00:13:37.655
than a high volatility security may add more vol,

286
00:13:37.655 --> 00:13:41.855
value at risk than a, a low volatility security in terms

287
00:13:41.855 --> 00:13:43.575
of, return volatility, for example.

288
00:13:44.675 --> 00:13:48.495
So we also, estimated these heterogeneous,

289
00:13:48.775 --> 00:13:51.015
responses across different types of security and,

290
00:13:51.015 --> 00:13:52.495
and the findings really line up.

291
00:13:52.635 --> 00:13:54.335
We find those,

292
00:13:54.525 --> 00:13:57.295
that those value at risk shifts really matter more

293
00:13:57.355 --> 00:13:59.895
for securities that are risk in the sense of

294
00:13:59.895 --> 00:14:02.735
that they have a relatively high return volatility.

295
00:14:02.835 --> 00:14:04.255
And for example, we don't see

296
00:14:04.605 --> 00:14:08.135
that value at risk constraints really impact the positions

297
00:14:08.135 --> 00:14:10.255
or the turnover or the spread of TBIs,

298
00:14:10.255 --> 00:14:12.655
which are really short beta treasury securities,

299
00:14:12.655 --> 00:14:17.095
which are typically, seen as not that risky compared

300
00:14:17.095 --> 00:14:18.455
to longer term securities.

301
00:14:18.455 --> 00:14:20.255
So this lines up very nicely with the,

302
00:14:20.405 --> 00:14:22.335
with the mechanism that we have in mind.

303
00:14:22.775 --> 00:14:25.455
Treasury markets are at the heart of the financial system,

304
00:14:25.605 --> 00:14:27.335
both in the United States and globally.

305
00:14:28.155 --> 00:14:32.695
Um, so, from a regulatory perspective,

306
00:14:32.925 --> 00:14:36.495
it's important to understand the impact of,

307
00:14:37.455 --> 00:14:42.335
regulatory constraints, that target either, bank,

308
00:14:42.515 --> 00:14:43.615
balance sheets

309
00:14:43.955 --> 00:14:48.655
or, their, intermediation capacity,

310
00:14:48.915 --> 00:14:50.415
via, via the training desks.

311
00:14:50.695 --> 00:14:53.445
Ultimately, this, intermediation capacity we show is

312
00:14:53.445 --> 00:14:56.445
crucial, for treasury market liquidity more broadly,

313
00:14:56.785 --> 00:15:00.405
our results focus on the fact that there's sort

314
00:15:00.405 --> 00:15:04.485
of just this, this set, granular, smaller set

315
00:15:04.545 --> 00:15:05.765
of, key dealers.

316
00:15:06.105 --> 00:15:08.845
Um, that's, that's really critical for, treasury,

317
00:15:09.105 --> 00:15:11.085
intermediation and treasury functioning.

318
00:15:11.665 --> 00:15:16.525
Um, in terms of sort of the impact of treasury markets on,

319
00:15:16.625 --> 00:15:21.605
on policy, you know, monetary policy,

320
00:15:21.905 --> 00:15:23.565
transmission,

321
00:15:23.835 --> 00:15:26.605
obviously very much relies on treasury market,

322
00:15:27.485 --> 00:15:31.365
functioning and in an era of increased, government debt.

323
00:15:31.785 --> 00:15:35.525
Um, the ability of banks to intermediate this,

324
00:15:35.595 --> 00:15:38.725
this debt market is, is really crucial.

Impact of Primary Dealers' Constraints on Treasury Market Liquidity

00:00:00 - The idea that treasuries play a crucial role in the global economy. They're the primary means of financing the US Federal Government. they're a significant investment instrument in hedging vehicle for global investors. They're risk-free benchmark for many other financial institutions, and they're an important tool for the Federal Reserves implementation of monetary policy. that said, for how important they are, they're intermediated by just a select group of market making financial firms, which are called primary dealers. and these primary dealers are typically affiliated

00:00:32 - So the question was, and the idea was, what happens when these primary dealers face, constraints? How does that impact the treasury market? And our work shows that when, the primary dealers are faced with tighter constraints, and we can sort of talk about, defining what we mean by constraints. these primary dealers do indeed reduce their treasury holdings, and this is accompanied by lower liquidity in the treasury market.

00:01:02 - Work really, is, is, is grounded in this relatively large literature, especially theoretical literature on the role of financial intermediaries, in asset pricing and specifically, there's a relatively large literature that focuses on what constraint to these constraints to these intermediaries really do, in terms of asset pricing and liquidity. So part of the literature, was really inspired by the events of the great financial crisis.

Empirical Insights on Financial Intermediaries Constraints

00:01:33 - mostly theoretical papers in the aftermath. therefore, but, there's also a host of papers that tries to get at these questions empirically. I'm trying to find empirical support for the theories that, that argue that financial intermediaries constraints really are important for driving asset prices. What we add relative to the existing, literature is really that we have a very unique, very detailed, granular data on individual actors

00:02:04 - So we're really able to go really into the nitty gritty details and really disentangle how those, constraints that these primary dealer face really, drive their decisions in terms of, treasury holdings, in terms of, turnover, in terms of, spread they charge. So we are really, adding to this literature by providing these very detailed micro data that allows us to costly identify, these effects. We use, two, supervisory,

00:02:38 - data sets, one of which is, supervisory. We use, the FR 2004, which is the government securities dealer reports. And we use the, regulation VV quantitative, measurements, which we call the, the VP one, dataset. So the, FR 2004, that's a, sort of longer historical data set, that includes information on the outright positions and the turnover of all primary dealers

Analysis of Trading Desk Data and Risk Constraints

00:03:08 - This data set includes, the important response variables that I talked about earlier, of, positions and turnovers of these primary dealers. The second data set that we use it, the FFR VV one provides desk specific internal, risk limits for the trading desks of large banking organizations. So, this data set is relatively newer. It's been collected since July, 2014,

00:03:44 - The VUCA rule prohibits, banks from engaging in proprietary trading, but as part of that rule, banks are required to report their internal organization of trading desks, desk names, descriptions, and also report the, internal risk limits, of these trading desks. In practice. These risk limits take a couple forms. they can be sensitivity of limits, they can be value at risk or VAR limits. And we primarily focus on the VAR limits.

00:04:14 - with our Y nine C, and then, auction data as well as, some pricing data. The question is, what happens, when banks face tighter constraints? what happens to their positions? What happens to turnover? What happens to the best spreads and treasury market liquidity? so, the, I think let's, we can back up for a second and ask the question of what do we mean by constraints? so banks, have sort of a variety.

Understanding Banking Constraints on Treasury Trading

00:04:47 - which is composed of different organizational layers, and you can imagine constraints at different points of the banking structure. You can imagine constraints on the, overall bank balance sheet or, especially relevant for these, primary dealers which are trading these, the government securities, on the actual trading desk themselves. and so because banks can face constraints in different places, we actually use two different methods to

00:05:17 - of two different types of constraints. so first we start with the SLR constraint, which, is part of the US implementation of the Basel three accords. and the idea here is, the SLR is the ratio of tier one capital to total leverage exposure, because exposure includes gross positions in securities, held by primary dealers. SLR constrained banks might be constrained in their treasury positions.

00:05:47 - and we're gonna use a particular method to, to explore the impact of constraint one, constraint two is these internal constraints on banks trading activities. and so as I mentioned, the ULCA rule, does limit banks from taking too much risk and engaging in proprietary trading. But separate from that, even though treasuries aren't, they're exempt from the, from the ULCA rule, banks still have internal risk limits

Analyzing SLR and VR Constraints on Treasury Trading

00:06:21 - you know, unlimited risk and, and treasuries. And we actually can see in our data that there are, value at risk constraints on the trading desk that only trade treasuries. And so we have a second method for looking at the impact of these, internal, VR risk limits on the, on the training desk. When we look at the SLR, the impact of SLR constraints, we use a difference in difference methodology, which leverages the fact that in April, 2020,

00:06:57 - of this SLR constraint, in which treasuries were exempt from the SLR. and so what we do is we compare, banks that were more constrained or less constrained in their SLR according to the 2019, Q4 value of their SLR. We compare, how more constrained banks reacted after this,

00:07:25 - less constrained banks. so that's the first methodology, and, that allows us to look at sort of differences in positions between more constrained and less constrained banks when that is relaxed. our second, method, uses the granular instrumental, or something sort of similar to the granular instrumental variable approach. And the idea here is that if only,

Analyzing Exogenous Risk Limit Changes in Treasury Markets

00:07:58 - that risk limit change is, probably exogenous to overall market conditions. And so you can look at these sort of, these risk limit changes that are only happening at one desk, and you do that by running this, saturated fixed effects regression. and then you aggregate all of these idiosyncratic changes up, and that gets this exogenous,

00:08:29 - and then we can sort of regress this exogenous risk limit shock, or look, look, regress some positions and, turnover and, bitas spreads on this exogenous risk limit shock Looking essentially as at, at two depths of analysis here, like related to the different constraints and related the different methods here. If we're, I'm gonna first talk about the findings for the

00:08:59 - So here we really exploit this, this policy change, in the wake of the covid, pandemic in early 2020 when the Federal Reserve, made a temporary, change to the SLR computation, in the following sense that, treasury exposure and, and also reserves held at the central banks were, as of April 1st, 2020 20, allowed

Impact of Policy Change on Treasury Holdings and Liquidity

00:09:30 - So what does it mean? It means that if a primary dealer affiliated with a bank holding company, holds a treasury, position in its book, that treasury position will not add to the leverage exposure, therefore will not be costly in terms of capital. And that change really was, announced like on a very short notice and became effective immediately. So we're looking around this, tight window of a couple of, weeks around before and after that policy change,

00:10:03 - that change in terms of, holdings and, and turnover and spread. And so what we really find is like, very intuitive if this, constraint was lifted by exempting the treasury securities from the total leverage exposures, primary dealers that are affiliated with a bank holding company that has a relatively low leverage ratio that is relatively more constrained than other, other,

00:10:34 - increase their position in treasuries after the, after the policy change, and they increase also their total turnover. And we also see that their, margin, their kind of a measure of intermediation spread that that OM declines essentially. I can also put some precise numbers on it. These effects are relatively sizable. So our estimates indicate that, primary dealer associated with a bank

Impact of Leverage Ratio Changes on Treasury Market Liquidity

00:11:04 - leverage ratio, a one percentage point lower leverage ratio, after the policy policy change, those primary dealers, see an increase in about nine percentage points relative to other dealers in their pro treasury position holdings. We also see a, higher turnover and a, and a and a lower, bit are spread here. So this is all, according to the theory that these constraints matters. And once you lift those binding constraint,

00:11:35 - and a, a lower intermediation spread. So the, the market becomes more liquid in some sense. It's what is very interesting, is that we also looked at heterogeneous effect across different treasury securities. And our findings here show that there's actually no heterogeneous effects. After the policy change was enacted, the dealers increased their position holdings across all

00:12:04 - and also longer dated, securities. Given the, the, the regulatory framework. This is very intuitive because these supplementary leverage ratio, as Hillary mentioned, this is the, implementation in the US of the basal accords. This is meant to be risk insensitive. So we are finding exactly, this risk insensitive, effects. When we look at the, what, what the, temporary exemption of treasuries that to the, to the holdings, the primary dealers increase the holdings across the board

Impact of Value at Risk on Treasury Securities

00:12:36 - Where we look at the, at the supplementary leverage ratio, if you look at the effects of value at risk, shifts, as Hillary explained the, the method already, we are finding qualitatively very similar results. What we find is that as value at risk limits become tighter, the, primary dealers reduce their positions in treasury securities. As a result, we also see, total turnover going down and,

00:13:11 - So this is like qualitatively very similar to what we find, for the supplementary leverage ratio. There is, one important difference that i I wanna highlight here, which is that the conceptually, the value at risk constraint, as the name already suggests this, are risk sensitive constraints. So a security with where may add different value at risk, than a high volatility security may add more vol, value at risk than a, a low volatility security in terms

00:13:44 - So we also, estimated these heterogeneous, responses across different types of security and, and the findings really line up. We find those, that those value at risk shifts really matter more for securities that are risk in the sense of that they have a relatively high return volatility. And for example, we don't see that value at risk constraints really impact the positions or the turnover or the spread of TBIs, which are really short beta treasury securities,

Understanding Treasury Markets and Regulatory Impact

00:14:17 - to longer term securities. So this lines up very nicely with the, with the mechanism that we have in mind. Treasury markets are at the heart of the financial system, both in the United States and globally. so, from a regulatory perspective, it's important to understand the impact of, regulatory constraints, that target either, bank, balance sheets

00:14:48 - via, via the training desks. Ultimately, this, intermediation capacity we show is crucial, for treasury market liquidity more broadly, our results focus on the fact that there's sort of just this, this set, granular, smaller set of, key dealers. that's, that's really critical for, treasury, intermediation and treasury functioning. in terms of sort of the impact of treasury markets on,

00:15:35 - transmission, obviously very much relies on treasury market, functioning and in an era of increased, government debt. the ability of banks to intermediate this, this debt market is, is really crucial.

Faculti AI can make mistakes

Research

Policy

Analysis