-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy path02-lecture.Rmd
338 lines (231 loc) · 13.5 KB
/
02-lecture.Rmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
---
title: "Lecture 2: Colonialism and Sovereign Debt"
author: |
| Dr. Gabriel Geisler Mesevage
| Office Hours: Tues. 11am & Weds. 11am
date: "Last update: `r format(Sys.time(), '%d %B, %Y')`"
output:
xaringan::moon_reader:
nature:
beforeInit: "helper_functions/macros.js"
ratio: '16:9'
lib_dir: libs
css: ["xaringan-themer.css", "helper_functions/extra.css"]
---
```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = FALSE, warning = FALSE,
message = FALSE, fig.align='center', fig.retina=3,
out.width="75%")
```
```{r xaringan-themer, include = FALSE}
library(xaringanthemer)
style_solarized_light()
source("helper_functions/theme_lecture.R")
xaringanExtra::use_webcam()
xaringanExtra::use_tile_view()
```
## Introduction
.Large[### Today's Plan
+ Colonialism and Sovereign Debt: framing the problem
+ The return to institutions
+ Measuring institutions
+ Institutions in cross-country regressions
+ Colonialism and sovereign debt reconsidered
+ Accominotti et al
+ Gardner
]
---
## Colonies, Institutions and Growth
+ A resurgent literature on institutions and growth since the 1990s/200s
- Key contributions by Acemoglu et al take colonies as a source of variation in institutional form
- Disease burden $\rightarrow$ type of colony $\rightarrow$ type of institution $\rightarrow$ growth
+ Led to broader study of *effects of colonialism* on growth
- Obstfeld & Taylor (2003): Empire itself does little, gold standard is what matters
- Ferguson and Schularick (2006): Empire lowers borrowing costs
+ Imposes 'fiscal orthodoxy' onto imperial subjects
+ This potentially includes broad set of imperial institutions (e.g. gold standard, free trade)
### Is this a 'benefit' of empire?
> "apologists for Europe's imperial record over this period" <br> - (O'Brien 2006)
---
### An older literature on costs of empire
.pull-left40[
#### Davis and Huttenback, *Mammon and the Pursuit of Empire*
]
.pull-right60[
+ The strong case for the economic irrationality of empire
- Compute costs and revenues
- Revenues go to narrow elite
- Costs fall on the public: defense, administration, infrastructure, opportunity costs
- 'Winners' are specific elites and settler colony emigres
- 'Losers' are the British taxpayer and (obviously) the colonized
> "... if I invest either in the public funds or in some private industrial venture in a foreign country for the benefit of my private purse, getting specially favorable terms to cover risks arising from the political insecurity of the country or the deficiencies of its Government, I am entitled to call upon my Government to use its political and military force to secure those very risks which I have already discounted in the terms of my investment. Can anything be more palpably unfair?" <br> —Hobson, 1902, p. 358.
]
---
## Empire's winners and losers
.center.Large[
> "The technology of the imperial machine ... involved some transfer of resources to the colonies; however, it is not obvious that either India or **the dependent colonies** [Ceylon, Jamaica, etc.] would have chosen to accept the imperial subsidy had they been given the opportunity to object ... **The colonies with responsible government** [Canada, Australia...] were clear winners; India and the dependent Empire, probably, were losers. ... The value of those benefits was high in the colonies of white settlement ... For the remainder of the Empire, the returns are less obvious ... For India and the dependent colonies one cannot rule out the conclusion that everyone (Briton and Indian) lost – a true Pareto pessimum." <br> - Davis and Huttenback quoted in Accominatti et al. 2009 fn. 5.
]
---
## The econometrics of empire: the conventional approach
+ How to measure the average impact of empire on borrowing costs?
$Spread_{ct} = \beta_1 Int_{ct} + \beta_2 Budg_{ct} + \beta_3 Trade + \beta_4 Default_{ct} + \beta_5 Prev.Default_{ct} + C_c + T_t + \epsilon_{ct}$
+ $Spread_{ct}$ is the difference between the bond yield of country $c$ and UK consols at time $t$
- consols are here considered the 'risk-free rate'. The Spread will in general always be positive as everyone pays more to borrow than UK gov
+ The parameters $C_c$ are the average increase/decrease in borrowing costs associated with each country $c$
- If $C_c$ is **negative** then Spreads are lower for country $c$
.pull-left[
#### Testing for the empire effect
+ Countries fall in two groups $C_c \in \{Colony, Sovereign\}$
+ If $\bar{C}_{Colony} < \bar{C}_{Sovereign}$ where $\bar{X}$ means the average of $X$ than on average colonies borrow at lower rates
- Why not include a variable for colony?
]
.pull-right[
|Country | Year | Colony
|-------------------------
|India | 1900 | 1
|India | 1901 | 1
| ... | ... | ...
|France | 1900 | 0
|France | 1900 | 0
]
---
.pull-left60[
### What are we assuming?
Say we want to estimate how the average of $Y$ changes with $X$ in a linear model
$Y_{ct} = \alpha + \beta X_{ct} + \epsilon_{ct}$
The problem is that we do not observe the 'errors' $\epsilon_{ct}$ and they might vary with $X$ and $Y$,
+ e.g. $Y$ is GDP growth, $X$ is a measure of industrialization, but both are explained by institutions $I$ which we can't measure, and global weather patterns $W$ which we can't measure?
+ What if the things we can't measure only vary **either** across countries **or** across time?
$Y_{ct} = \alpha + \beta X_{ct} + I_c + W_t + u_{ct}$
Now if you estimate the average for each country, and you estimate the average for each year, you have removed the effects of $I$ and $W$ **and anything else that varies only by place or time** even though you haven't measured them!
]
.pull-right40[
### Is this useful?
+ For the purposes of measuring $\beta$ we've removed a lot of tricky unobservables
- But hardly a silver bullet: anything we didn't measure that changes across space and time is a problem (education, population, soil quality, whatever)
### What about testing for empire?
The empire test is **using** the country averages it estimates. Imagine:
$Y_{ct} = \alpha + \beta X_{ct} + Empire_c + I_c + W_t + u_{ct}$
+ Effect of empire would be **unidentifiable** in this model
]
---
## The econometrics of empire
.pull-left[
]
.pull-right[
+ Ferguson & Schularick, p. 299 table 3.
+ Colonies borrow ~1.5% cheaper than sovereigns!
]
---
.pull-left[
## The econometrics of empire
### But the result is unstable!
+ Obstfeld & Taylor ran a similar regression and got a different result
+ Ferguson and Schularick suggest their sample is larger
- But many reasons for unstable result, most of them frustrating:
+ Measurement error, omitted-variable bias, **mispecification**...
]
.pull-right[
]
---
## How did investors think about country risk?
+ Argue most of the econometric approaches have been **mispecified** because the pricing model is different for **colonies** versus **sovereigns**.
+ **sovereigns** have a probability of default that sharply increases as their revenues fall, trade falls, etc.
+ But **colonies** are implicitly/explicitly guaranteed by colonizer. They don't have the **same** kind of default risk. Spreads represent some little risks of default + other issues like liquidity
---
.left-column[
## Example
+ Colonies less sensitive to debt burden than sovereigns
+ Logic is that we should run 2 separate regressions
]
.right-column[
.center.middle[
]]
---
## The econometrics of empire re-specified
+ Need to think in terms of seperate relationships between spreads and 'fundamentals' for colonies and non-colonies
.pull-left[
+ 1 equation for colonies
$Spread_{ct}^{COL} = \beta_1^{COL} Int_{ct} + \beta_2^{COL} Budg_{ct} + \\ \beta_3^{COL} Trade + \beta_4^{COL} Default_{ct} +\\ \beta_5^{COL} Prev.Default_{ct} + C_c^{COL} + T_t^{COL} + u_{ct}$
+ And a seperate equation for sovereigns
$Spread_{ct}^{SOV} = \beta_1^{SOV} Int_{ct} + \beta_2^{SOV} Budg_{ct} +\\ \beta_3^{SOV} Trade + \beta_4^{SOV} Default_{ct} +\\ \beta_5^{SOV} Prev.Default_{ct} + C_c^{SOV} + T_t^{SOV} + v_{ct}$
]
.pull-left[
]
---
## Econometrics and interpretations
.pull-left[
> "The Empire made the spread paid by subject countries insensitive to their performance because credibility was decided elsewhere. You would not look in India for indications of India's credit. More likely, you would look at Downing Street. The effect of empire was not to provide subjects with a marginal interest rate benefit but to remove the default risk altogether" <br> - Accominotti et al. p. 399
]
.pull-right[
+ Does not make sense to use the same pricing model for India and Argentina when only one of them is sovereign!
- Similar to mixing Alabama and Argentina in a study
+ **If colonies can't default, what limits their borrowing?**
- Politics
+ **If colonies don't have default risk, not sensible to think in terms of their macroeconomic policies**
]
---
## What determines colonial borrowing?
.pull-left[
> "The management of capital outflows within the British Empire was thus tantamount to redistributive policies in federal regimes - if political science is any guide here, the crucial variable for determining the resulting patterns is the degree of devolution." <br> -Accominotti et al. p. 400
]
.pull-right[
]
+ **White settler colonies** become increasing indebted across later-19th
+ Reflects their greater political leverage within the Empire
- Politics is conditioning capacity to borrow
---
### What does this tell us about developmental strategies?
> Our theoretical analysis leads us to predict that **the institutions of self-governing colonies were conducive to heavier government involvement in the process of economic growth than was the case under sovereignty**. We report empirical evidence that is consistent with this. This finding is at odds with the prejudice recently associated with neoconservative writing, that empire is useful because its repressive nature is conducive to financial orthodoxy. ...certain countries such as Australia and New Zealand were pioneers in the development of welfare state institutions. We remark that they belonged to the self-governing empire and suggest that their pioneering of the welfare state was a by-product of the political economy of self-governing colonies, where extensive government involvement in economic development and a deepening of the electoral franchise went hand in hand.
.pull-left[
]
.pull-right[
+ Analyzing policies/institutions in isolation from the interests of colonial ruler is problematic
+ Self-governing (white settler e.g. Australia, Canada, etc.) colonies need to be cajoled
- Get subsidies, developmental assistance, etc.
+ Dependent colonies (e.g. India, Jamaica, etc.) were coerced
]
---
### What do case studies suggest?
.center[
]
+ Recall the difficulty and ambiguity of the statistical work
+ Garner proposes evaluating debate in context of West African case studies:
- Do investors lend because imperial institutions "made colonies a better credit risk" (p. 236): Obstfeld & Taylor; Ferguson and Schularick?
- Or "not because colonialism improved their fiscal and economic institutions but *because* they were colonies" (p. 400): Accominotti et al
+ Nigeria, Sierra Leone and Gold Coast (colonies) vs Liberia (Independent)
---
## What does Gardner find? History is messy
.pull-left[
> "At the core of the debate is whether imperial policies improved those fundamentals by enforcing fiscal discipline, or whether the status of the colonies as subsidiaries of the British government explains the low level of spreads for colonial bonds." <br> -Gardner, p. 254
+ I read her as largely siding with Accominotti et al.
+ **But** colonials *do* pay a spread above London -- where does it come from?
- Gardner suggests some real investor fears about revenue/default
- Chavaz and Flandreau (2017, *Journal of Economic History*) argue it is liquidity
- (could be both)
]
.pull-right[
> "The experience of West African colonies suggests a more nuanced approach is needed to this debate, which allows for the fact that investors did appear to differentiate between colonial bonds. A wide range of imperial interventions, which included not only the Crown Agents but also the British government itself, contributed to allowing West African colonies to borrow on comparatively good terms even relative to other British colonies. Such interventions served the purposes of various interests within the British Empire and need not necessarily be attributed to benevolence. <br> -Gardner, p. 254
]
---
## Concluding: some things to consider
.Large[
+ What should we make of the inverse relationship between 'fiscal orthodoxy' and growth?
+ How much stock should we put in cross-country regressions?
+ Empire functioned to subsize some colonies and not others
- How should this figure in our accounts of growth?
- How problematic is this for accounts grounded in *good* vs *bad* institutions?
+ What does this tell us about the *origins* of institutional forms in the 19th century?
+ What can sovereign borrowing costs tell us about interesting questions in econ history/political economy?
]