Impacts of EU Sanctions Levied in 2014 on Individual European Countries' Exports to Russia: Winners and Losers

Abstract

We analyse the effects of sanctions implemented by the European Union against Russia following the latter’s annexation of Crimea in 2014. Indirect effects of sanctions on its non-prohibited exports to Russia are examined using a gravity model of trade that includes a time varying sanction index. A per country analysis is also incorporated to increase the granularity of the results. We find that sanctions led to a decrease in exports of non-prohibited products from certain European countries (i.e., the “losers”) while increasing such exports from others (i.e., the “winners”), an outcome that qualifies as an “unintended consequence” of the sanctions.

Appendices

Appendices

Due to their excessive length, appendices A, L, M, O, P, and Q, are all available online at https://www.moradbali.com/publications. For more information or requests, feel free to write to morad.bali@irnc.org.

Appendix B: European country code and their attributed numbers

Code	Name	Attributed number (i)	Code	Name	Attributed number (i)
AT	Austria	1	HU	Hungary	15
BE	Belgium	2	IE	Ireland	16
BG	Bulgaria	3	IT	Italy	17
CY	Cyprus	4	LT	Lithuania	18
CZ	The Czech Republic	5	LU	Luxembourg	19
DE	Germany	6	LV	Latvia	20
DK	Denmark	7	MT	Malta	21
EE	Estonia	8	NL	Netherlands	22
ES	Spain	9	PO	Poland	23
FI	Finland	10	PT	Portugal	24
FR	France	11	RO	Romania	25
GB	The United Kingdom	12	SE	Sweden	26
GR	Greece	13	SI	Slovenia	27
HR	Croatia	14	SK	Slovakia	28

Appendix C: Data sources

Variable	Details
${\text{XRU}}_{t,i}$	Countries’ Exports to Russia—Eurostat; extracted on 03.06.20; last update 15.05.20; total exports net of arms and ammunition (SITC 891); value in euros; EU trade since 1988 by SITC [DS-018995]. Products targeted by the Russian embargo were also removed from this database.
${\text{LP}}_{t,i}$ 1	Labour Productivity[1] (per hours worked)—European Central Bank; total economy; all activities; index; chain linked volume (rebased); non-transformed data; neither seasonally adjusted nor calendar adjusted data; average of observations through period (A).
${\text{PPI}}_{t,i}$	Producer Prices in Industry—Eurostat; extracted on 04.06.20; last update 03.06.20; total—quarterly data; total output price index—in national currency; industry (except construction, sewerage, waste management and remediation activities); unadjusted data (i.e. neither seasonally adjusted nor calendar adjusted data); index, 2015=100, [sts_inpp_q].
${\text{REER}}_{t,i}$	Real Effective Exchange Rate—Eurostat; extracted on 04.06.20; last update 03.06.20; deflator: consumer price index—42 trading partners—industrial countries; Index, 2010=100; [ert_eff_ic_q].
${\text{ER}}_{t,i}$	Country’s Currency Exchange Rate Against the Russian rouble—European Central Bank; ECB reference exchange rate; average of observations through period (A).
${\text{GDP}}_{t,r}$	Russia’s GDP—Organization for Economic Co-operation and Development; gross domestic product by expenditure in constant prices: total gross domestic product for the Russian Federation [NAEXKP01RUQ652S], retrieved from FRED, Federal Reserve Bank of St. Louis.
${\text{HICP}}_{t,i}$	Harmonised Index of Consumer Prices—Eurostat; extracted on 04.06.20; last update 29.05.20; index, 2015=100; neither seasonally adjusted nor calendar adjusted data; all items; monthly data [ei_cphi_m].
${\text{ML}}_{t,i}$	ECB Marginal lending facility—European Central Bank; euro area (changing composition); key interest rate; date of changes (raw data); level; euro; average of observations through period; per cent per annum.
${\text{S}}_{t}$	Sanction Index (European Sanctions)—Bali and Rapelanoro (2021).
${\text{distance}}_{ij}$	Distance between capitals of European countries and Russia; extracted from the CEPII Mayer and Zignago (2011).

[1] Regarding ${\text{ LP}}_{t,i}$, data for Belgium, Greece, Luxembourg and the United Kingdom were not available as detailed here. Thus, it was necessary to use instead:

• Greece: Labour Productivity (per hours worked)—Greece—world (all entities, including reference area, including io), total economy, services, index, chain linked volume (rebased), non-transformed data, neither seasonally adjusted nor calendar adjusted data, ECB.

• Luxembourg: Labour Productivity (per persons)—Luxembourg—world (all entities, including reference area, including io), total economy, total—all activities, index, chain linked volume (rebased), non-transformed data, neither seasonally adjusted nor calendar adjusted data, ECB.

• The United Kingdom: Labour Productivity (per hours worked)—United Kingdom—world (all entities, including reference area, including io), total economy, services, index, chain linked volume (rebased), non-transformed data, neither seasonally adjusted nor calendar adjusted data, ECB.

• Belgium: Labour Productivity (per persons)—Belgium—world (all entities, including reference area, including io), total economy, total—all activities, index, chain linked volume (rebased), non-transformed data, neither seasonally adjusted nor calendar adjusted data, ECB.

[2] Regarding ${\text{PPI}}_{t,i}$, data are not available for Ireland and Portugal between 2003 and 2004. Consequently, there are missing 16 observations (8 quarters for each country).

Appendix D: Impact of EU countries’ sanctions on export to Russia: the raw sanction values

	Export of European countries to Russia
	Model (1.1)	Model (1.2)	Model (1.3)	Model (1.4)
European sanctions index	−0.000	0.000	−0.000	0.000
	(0.000)	(0.000)	(0.000)	(0.000)
GDP of the RF (ln)	1.861***	1.635***	1.435***	1.456***
	(0.227)	(0.267)	(0.241)	(0.229)
GDP of the EU countries (ln)	0.873***	0.873***	0.893***	0.892***
	(0.072)	(0.075)	(0.085)	(0.083)
Distance between capitals (ln)	−2.003***	−1.993***	−1.922***	−1.917***
	(0.294)	(0.293)	(0.292)	(0.285)
Real effective exchange rate	−0.011		−0.000
	(0.012)		(0.012)
Harmonised index of consumer prices	0.007	0.006
	(0.007)	(0.006)
Marginal lending key interest rate	−0.072*	−0.046	−0.080**	−0.069*
	(0.037)	(0.036)	(0.040)	(0.041)
Labour productivity	0.013***	0.009**	0.011**	0.010**
	(0.004)	(0.004)	(0.005)	(0.005)
Exchange rate against the Russian rouble		0.019***		0.014
		(0.006)		(0.021)
Producer prices in industry			0.010	0.002
			(0.009)	(0.016)
Sanction time (Dummy)	0.250***	0.071	0.199***	0.129
	(0.085)	(0.068)	(0.066)	(0.096)
_cons	−0.233	0.185	2.908	2.081
	(2.685)	(3.181)	(2.658)	(2.829)
Number of observations	1904	1904	1888	1888
R-squared	0.889	0.889	0.873	0.878

1. Robust standard errors clustered at country level in parentheses; ln: natural logarithm; In model (1.3) and (1.4), 16 observations are excluded due to the missing data in producer prices in industry (the total observations decrease from 1904 to 1888); ^∗∗∗ p<0.01, ^∗∗ p<0.05, ^∗ p<0.1

Appendix E: Impact of EU countries’ sanctions on export to Russia: the lagged sanction values

	Export of European countries to Russia
	Model (2.1)	Model (2.2)	Model (2.3)	Model (2.4)
European sanctions index	−0.000	0.000	0.000	0.000
	(0.000)	(0.000)	(0.000)	(0.000)
GDP of the RF (ln)	1.848***	1.642***	1.434***	1.461***
	(0.225)	(0.263)	(0.234)	(0.230)
GDP of the EU countries (ln)	0.873***	0.873***	0.893***	0.892***
	(0.072)	(0.075)	(0.085)	(0.083)
Distance between capitals (ln)	−2.003***	−1.993***	−1.922***	−1.917***
	(0.294)	(0.293)	(0.292)	(0.285)
Real effective exchange rate	−0.012		−0.000
	(0.012)		(0.012)
Harmonised index of consumer prices	0.007	0.006
	(0.007)	(0.006)
Marginal lending key interest rate	−0.071*	−0.046	−0.080**	−0.069*
	(0.037)	(0.036)	(0.040)	(0.040)
Labour productivity	0.013***	0.009**	0.011**	0.010**
	(0.004)	(0.004)	(0.005)	(0.005)
Exchange rate against the Russian rouble		0.019***		0.014
		(0.006)		(0.021)
Producer prices in industry			0.010	0.002
			(0.009)	(0.016)
Sanction time (Dummy)	0.241***	0.075	0.193***	0.128
	(0.083)	(0.066)	(0.063)	(0.090)
_cons	−0.063	0.100	2.912	2.013
	(2.662)	(3.127)	(2.651)	(2.831)
Number of observations	1904	1904	1888	1888
R-squared	0.889	0.889	0.873	0.878

1. Robust standard errors clustered at country level in parentheses; ln: natural logarithm; In model (2.3) and (2.4), 16 observations are excluded due to the missing data in producer prices in industry (the total observations decrease from 1904 to 1888); ^∗∗∗ p<0.01, ^∗∗ p<0.05, ^∗ p<0.1

Appendix F: Impact of EU countries’ sanctions on export to Russia: the logarithm of the raw values

	Export of European countries to Russia
	Model (3.1)	Model (3.2)	Model (3.3)	Model (3.4)
European sanctions index	−0.007	0.005	−0.000	0.003
	(0.005)	(0.005)	(0.007)	(0.006)
GDP of the RF (ln)	1.860***	1.631***	1.433***	1.452***
	(0.227)	(0.268)	(0.247)	(0.233)
GDP of the EU countries (ln)	0.873***	0.873***	0.893***	0.892***
	(0.072)	(0.075)	(0.085)	(0.083)
Distance between capitals (ln)	−2.003***	−1.993***	−1.922***	−1.916***
	(0.294)	(0.293)	(0.292)	(0.285)
Real effective exchange rate	−0.011		−0.000
	(0.012)		(0.012)
Harmonised index of consumer prices	0.007	0.006
	(0.007)	(0.006)
Marginal lending key interest rate	−0.072*	−0.045	−0.080**	−0.068*
	(0.037)	(0.037)	(0.040)	(0.041)
Labour productivity	0.013***	0.009**	0.011**	0.010**
	(0.004)	(0.004)	(0.005)	(0.005)
Exchange rate against the Russian rouble		0.019***		0.014
		(0.006)		(0.021)
Producer prices in industry			0.010	0.002
			(0.010)	(0.016)
Sanction time (Dummy)	0.309***	0.022	0.198	0.101
	(0.095)	(0.099)	(0.122)	(0.139)
_cons	−0.260	0.239	2.928	2.124
	(2.682)	(3.201)	(2.695)	(2.874)
Number of observations	1904	1904	1888	1888
R-squared	0.889	0.890	0.873	0.878

1. Robust standard errors clustered at country level in parentheses; ln: natural logarithm; In model (3.3) and (3.4), 16 observations are excluded due to the missing data in producer prices in industry (the total observations decrease from 1904 to 1888); ^∗∗∗ p<0.01, ^∗∗ p<0.05, ^∗ p<0.1

Appendix G: Impact of EU countries’ sanctions on export to Russia: the logarithm of the lagged values

	Export of European countries to Russia
	Model (4.1)	Model (4.2)	Model (4.3)	Model (4.4)
European sanctions index	0.005	0.010***	0.011**	0.012***
	(0.006)	(0.003)	(0.004)	(0.002)
GDP of the RF (ln)	1.877***	1.689***	1.488***	1.515***
	(0.212)	(0.255)	(0.226)	(0.231)
GDP of the EU countries (ln)	0.873***	0.873***	0.893***	0.892***
	(0.072)	(0.075)	(0.085)	(0.083)
Distance between capitals (ln)	−2.003***	−1.992***	−1.922***	−1.917***
	(0.294)	(0.293)	(0.292)	(0.285)
Real effective exchange rate	−0.011		0.000
	(0.012)		(0.013)
Harmonised index of consumer prices	0.007	0.006
	(0.007)	(0.006)
Marginal lending key interest rate	−0.072*	−0.046	−0.080**	−0.069*
	(0.038)	(0.036)	(0.040)	(0.040)
Labour productivity	0.013***	0.009**	0.011**	0.010**
	(0.004)	(0.004)	(0.005)	(0.005)
Exchange rate against the Russian rouble		0.019***		0.014
		(0.006)		(0.021)
Producer prices in industry			0.010	0.002
			(0.009)	(0.016)
Sanction time (Dummy)	0.204***	0.007	0.114	0.048
	(0.053)	(0.064)	(0.077)	(0.091)
_cons	−0.454	−0.448	2.236	1.371
	(2.604)	(3.024)	(2.647)	(2.809)
Number of observations	1904	1904	1888	1888
R-squared	0.888	0.890	0.873	0.878

1. Robust standard errors clustered at country level in parentheses; ln: natural logarithm; In model (4.3) and (4.4), 16 observations are excluded due to the missing data in producer prices in industry (the total observations decrease from 1904 to 1888); ^∗∗∗ p<0.01, ^∗∗ p<0.05, ^∗ p<0.1

Appendix H: Impact of EU countries’ sanctions on export to Russia: the first difference of raw sanction values

	Export of European countries to Russia
	Model (5.1)	Model (5.2)	Model (5.3)	Model (5.4)
European sanctions index (first difference of sanction values)	−0.000	0.000	−0.000*	−0.000
	(0.000)	(0.000)	(0.000)	(0.000)
GDP of the RF (ln)	1.865***	1.639***	1.459***	1.474***
	(0.212)	(0.260)	(0.243)	(0.236)
GDP of the EU countries (ln)	0.873***	0.873***	0.893***	0.892***
	(0.072)	(0.075)	(0.085)	(0.083)
Distance between capitals (ln)	−2.003***	−1.993***	−1.922***	−1.917***
	(0.294)	(0.293)	(0.292)	(0.285)
Real effective exchange rate	−0.011		−0.000
	(0.012)		(0.012)
Harmonised index of consumer prices	0.007	0.006
	(0.007)	(0.006)
Marginal lending key interest rate	−0.072*	−0.046	−0.081**	−0.069*
	(0.038)	(0.036)	(0.040)	(0.041)
Labour productivity	0.013***	0.009**	0.011**	0.010**
	(0.004)	(0.004)	(0.005)	(0.005)
Exchange rate against the Russian rouble		0.019***		0.014
		(0.006)		(0.020)
Producer prices in industry			0.010	0.002
			(0.009)	(0.016)
Sanction time (Dummy)	0.240**	0.079	0.207***	0.140
	(0.097)	(0.069)	(0.063)	(0.092)
_cons	−0.302	0.148	2.600	1.864
	(2.572)	(3.079)	(2.720)	(2.905)
Number of observations	1904	1904	1888	1888
R-squared	0.888	0.890	0.873	0.878

1. Robust standard errors clustered at country level in parentheses; ln: natural logarithm; In model (5.3) and (5.4), 16 observations are excluded due to the missing data in producer prices in industry (the total observations decrease from 1904 to 1888); ^∗∗∗ p<0.01, ^∗∗ p<0.05, ^∗ p<0.1

Appendix I: Variance inflation factor (VIF) values for the estimation models with the logarithms of the raw values and lagged sanction values

	The case of logarithm of the raw values				The case of logarithm of the lagged values
	Model (3.1)	Model (3.2)	Model (3.3)	Model (3.4)	Model (4.1)	Model (4.2)	Model (4.3)	Model (4.4)
European sanctions index	10.51	10.73	10.62	10.69	8.42	8.37	8.38	8.35
GDP of the RF (ln)	3.78	4.15	4.37	4.37	4.11	4.48	4.71	4.71
GDP of the EU countries (ln)	1.17	1.14	1.16	1.15	1.17	1.14	1.16	1.15
Distance between capitals (ln)	1.27	1.25	1.13	1.12	1.27	1.25	1.13	1.12
Real effective exchange rate	1.29		1.13		1.30		1.13
Harmonised index of consumer prices	1.84	1.59			1.84	1.59
Marginal lending key interest rate	1.56	2.17	1.89	2.14	1.56	2.15	1.87	2.11
Labour productivity	1.60	1.73	1.70	1.74	1.57	1.71	1.68	1.72
Exchange rate against the Russian rouble		3.44		7.10		3.37		7.05
Producer prices in industry			2.03	4.18			2.01	4.17
Sanction time (Dummy)	14.53	17.05	15.90	16.95	9.00	10.29	9.67	10.24
Mean VIF	4.17	4.80	4.44	5.49	3.36	3.82	3.53	4.51

Appendix J: Variance inflation factor (VIF) values for the estimation models with the first difference of sanction values and the first difference of the logarithms of sanction values

	The first difference of sanction values				The first difference of the logarithms of sanction values
	Model (1)	Model (2)	Model (3)	Model (4)	Model (1)	Model (2)	Model (3)	Model (4)
European sanctions index	1.09	1.10	1.09	1.10	1.63	1.72	1.69	1.72
GDP of the RF (ln)	4.06	4.50	4.71	4.72	5.61	6.66	6.95	7.06
GDP of the EU countries (ln)	1.17	1.14	1.16	1.15	1.17	1.14	1.16	1.15
Distance between capitals (ln)	1.27	1.25	1.13	1.12	1.27	1.25	1.13	1.12
Real effective exchange rate	1.30		1.13		1.31		1.13
Harmonised index of consumer prices	1.84	1.59			1.86	1.60
Marginal lending key interest rate	1.56	2.15	1.87	2.12	1.56	2.22	1.91	2.18
Labour productivity	1.58	1.71	1.68	1.72	1.73	1.78	1.78	1.79
Exchange rate against the Russian rouble		3.39		7.09		3.60		7.21
Producer prices in industry			2.01	4.18			2.10	4.18
Sanction time (Dummy)	4.62	5.91	5.30	5.90	6.54	8.73	7.75	8.69
Mean VIF	2.05	2.53	2.23	3.23	2.52	3.19	2.84	3.90

Appendix K: Additional models

Different Options Explored to Cast the Sanction Index

We tried different options to select which form of the sanction index we will use: (i) raw values (i.e. untransformed) ; (ii) lagged values (t–1); (iii) logarithms of raw values; (iv) logarithms of lagged values (t–1); (v) the first difference of raw values; (vi) first difference of the logarithm of the sanction index. Forms (i-v) are not used in our analysis as they either lead to statistically insignificant results (Appendixes D-H) or suffer from multicollinearity (Appendixes I and J). Given that form (vi) is the only one that does not suffer from multicollinearity, we use it for the remainder of our analysis. Additionally, form (vi) is highly relevant for this analysis as there is often a delay in the coming into force of sanctions.

Models with Different Variable Combinations

In set (7), ${\text{HICP}}_{t,i}$ and ${\text{ER}}_{t,i}$ are removed, in set (8), ${\text{REER}}_{t,i}$ and ${\text{PPI}}_{t,i}$ are excluded, in set (9), ${\text{PPI}}_{t,i}$ and ${\text{ER}}_{t,i}$ are removed, and in set (10), ${\text{REER}}_{t,i}$ and ${\text{HICP}}_{t,i}$ are removed.

$$\begin{aligned} {\text{XRU}}_{t,i} = & \alpha_{c1} + \beta_{c1} {\text{FD}}(\ln \_S_{t} ) + \gamma_{c1} \ln \_{\text{GDP}}_{t,r} + \delta_{c1} \ln \_{\text{GDP}}_{t,i} + \theta_{c1} \ln \_{\text{dist}}_{r,i} \\ + & \;\vartheta_{c1} {\text{REER}}_{t,i} + \lambda_{c1} {\text{PPI}}_{t,i} + \varsigma_{c1} {\text{ML}}_{t,i} + \phi_{c1} {\text{LP}}_{t,i} + \psi_{c1} {\text{time}}_{t} + \varepsilon_{t,r,i}^{c1} \\ \end{aligned}$$

(7)

$$\begin{aligned} {\text{XRU}}_{t,i} = & \alpha_{b1} + \beta_{b1} S_{t} + \gamma_{b1} \ln \_{\text{GDP}}_{t,r} + \delta_{b1} \ln \_{\text{GDP}}_{t,i} + \theta_{b1} \ln \_{\text{dist}}_{r,i} \\ & \; + \rho_{b1} {\text{HICP}}_{t,i} + \varphi_{b1} {\text{ER}}_{t,i} + \varsigma_{b1} {\text{ML}}_{t,i} + \phi_{b1} {\text{LP}}_{t,i} + \psi_{b1} {\text{time}}_{t} + \varepsilon_{t,r,i}^{b1} \\ \end{aligned}$$

(8)

$$\begin{aligned} {\text{XRU}}_{t,i} = & \alpha_{a1} + \beta_{a1} S_{t} + \gamma_{a1} \ln \_{\text{GDP}}_{t,r} + \delta_{a1} \ln \_{\text{GDP}}_{t,i} + \theta_{a1} \ln \_{\text{dist}}_{r,i} \\ & \; + \vartheta_{a1} {\text{REER}}_{t,i} + \rho_{a1} {\text{HICP}}_{t,i} + \varsigma_{a1} {\text{ML}}_{t,i} + \phi_{a1} {\text{LP}}_{t,i} + \psi_{a1} {\text{time}}_{t} + \varepsilon_{t,r,i}^{a1} \\ \end{aligned}$$

(9)

$$\begin{aligned} {\text{XRU}}_{t,i} = & \alpha_{d1} + \beta_{d1} S_{t} + \gamma_{d1} \ln \_{\text{GDP}}_{t,r} + \delta_{d1} \ln \_{\text{GDP}}_{t,i} + \theta_{d1} \ln \_{\text{dist}}_{r,i} \\ & \; + \varphi_{d1} {\text{ER}}_{t,i} + \lambda_{d1} {\text{PPI}}_{t,i} + \varsigma_{d1} {\text{ML}}_{t,i} + \phi_{d1} {\text{LP}}_{t,i} + \psi_{d1} {\text{time}}_{t} + \varepsilon_{t,r,i}^{d1} \\ \end{aligned}$$

(10)

Sensitivity Tests

Models that use ($S_{t}^{a}$) instead of ($S_{t}$):

$$\begin{aligned} {\text{XRU}}_{t,i} = & \alpha_{c3} + \beta_{c3} {\text{FD}}(\ln \_S_{t}^{a} ) + \gamma_{c3} \ln \_{\text{GDP}}_{t,r} + \delta_{c3} \ln \_{\text{GDP}}_{t,i} + \theta_{c3} \ln \_{\text{dist}}_{r,i} \\ & \; + \vartheta_{c3} {\text{REER}}_{t,i} + \lambda_{c3} {\text{PPI}}_{t,i} + \varsigma_{c3} {\text{ML}}_{t,i} + \phi_{c3} {\text{LP}}_{t,i} + \psi_{c3} {\text{time}}_{t} + \varepsilon_{t,r,i}^{c3} \\ \end{aligned}$$

(11)

$$\begin{aligned} {\text{XRU}}_{t,i} = & \alpha_{b3} + \beta_{b3} {\text{FD}}(\ln \_S_{t}^{a} ) + \gamma_{b3} \ln \_{\text{GDP}}_{t,r} + \delta_{b3} \ln \_{\text{GDP}}_{t,i} + \theta_{b3} \ln \_{\text{dist}}_{r,i} \\ & \; + \rho_{b3} {\text{HICP}}_{t,i} + \varphi_{b3} {\text{ER}}_{t,i} + \varsigma_{b3} {\text{ML}}_{t,i} + \phi_{b3} {\text{LP}}_{t,i} + \psi_{b3} {\text{time}}_{t} + \varepsilon_{t,r,i}^{b3} \\ \end{aligned}$$

(12)

$$\begin{aligned} {\text{XRU}}_{t,i} = & \alpha_{a3} + \beta_{a3} {\text{FD}}(\ln \_S_{t}^{a} ) + \gamma_{a3} \ln \_{\text{GDP}}_{t,r} + \delta_{a3} \ln \_{\text{GDP}}_{t,i} + \theta_{a3} \ln \_{\text{dist}}_{r,i} \\ & \; + \vartheta_{a3} {\text{REER}}_{t,i} + \rho_{a3} {\text{HICP}}_{t,i} + \varsigma_{a3} {\text{ML}}_{t,i} + \phi_{a3} {\text{LP}}_{t,i} + \psi_{a3} {\text{time}}_{t} + \varepsilon_{t,r,i}^{a3} \\ \end{aligned}$$

(13)

$$\begin{aligned} {\text{XRU}}_{t,i} = & \alpha_{d3} + \beta_{d3} {\text{FD}}(\ln \_S_{t}^{a} ) + \gamma_{d3} \ln \_{\text{GDP}}_{t,r} + \delta_{d3} \ln \_{\text{GDP}}_{t,i} + \theta_{d3} \ln \_{\text{dist}}_{r,i} \\ & \; + \varphi_{d3} {\text{ER}}_{t,i} + \lambda_{d3} {\text{PPI}}_{t,i} + \varsigma_{d3} {\text{ML}}_{t,i} + \phi_{d3} {\text{LP}}_{t,i} + \psi_{d3} {\text{time}}_{t} + \varepsilon_{t,r,i}^{d3} \\ \end{aligned}$$

(14)

Models that use ($S_{t}^{b}$) instead of ($S_{t}$):

$$\begin{aligned} {\text{XRU}}_{t,i} = & \alpha_{c4} + \beta_{c4} {\text{FD}}(\ln \_S_{t}^{b} ) + \gamma_{c4} \ln \_{\text{GDP}}_{t,r} + \delta_{c4} \ln \_{\text{GDP}}_{t,i} + \theta_{c4} \ln \_{\text{dist}}_{r,i} \\ & \; + \vartheta_{c4} {\text{REER}}_{t,i} + \lambda_{c4} {\text{PPI}}_{t,i} + \varsigma_{c4} {\text{ML}}_{t,i} + \phi_{c4} {\text{LP}}_{t,i} + \psi_{c4} {\text{time}}_{t} + \varepsilon_{t,r,i}^{c4} \\ \end{aligned}$$

(15)

$$\begin{aligned} {\text{XRU}}_{t,i} = & \alpha_{b4} + \beta_{b4} {\text{FD}}(\ln \_S_{t}^{b} ) + \gamma_{b4} \ln \_{\text{GDP}}_{t,r} + \delta_{b4} \ln \_{\text{GDP}}_{t,i} + \theta_{b4} \ln \_{\text{dist}}_{r,i} \\ & + \rho_{b4} {\text{HICP}}_{t,i} + \varphi_{b4} {\text{ER}}_{t,i} + \varsigma_{b4} {\text{ML}}_{t,i} + \phi_{b4} {\text{LP}}_{t,i} + \psi_{b4} {\text{time}}_{t} + \varepsilon_{t,r,i}^{b4} \\ \end{aligned}$$

(16)

$$\begin{aligned} {\text{XRU}}_{t,i} = & \alpha_{a4} + \beta_{a4} {\text{FD}}(\ln \_S_{t}^{b} ) + \gamma_{a4} \ln \_{\text{GDP}}_{t,r} + \delta_{a4} \ln \_{\text{GDP}}_{t,i} + \theta_{a4} \ln \_{\text{dist}}_{r,i} \\ & \; + \vartheta_{a4} {\text{REER}}_{t,i} + \rho_{a4} {\text{HICP}}_{t,i} + \varsigma_{a4} {\text{ML}}_{t,i} + \phi_{a4} {\text{LP}}_{t,i} + \psi_{a4} {\text{time}}_{t} + \varepsilon_{t,r,i}^{a4} \\ \end{aligned}$$

(17)

$$\begin{aligned} {\text{XRU}}_{t,i} = & \alpha_{d4} + \beta_{d4} {\text{FD}}(\ln \_S_{t}^{b} ) + \gamma_{d4} \ln \_{\text{GDP}}_{t,r} + \delta_{d4} \ln \_{\text{GDP}}_{t,i} + \theta_{d4} \ln \_{\text{dist}}_{r,i} \\ & \; + \varphi_{d4} {\text{ER}}_{t,i} + \lambda_{d4} {\text{PPI}}_{t,i} + \varsigma_{d4} {\text{ML}}_{t,i} + \phi_{d4} {\text{LP}}_{t,i} + \psi_{d4} {\text{time}}_{t} + \varepsilon_{t,r,i}^{d4} \\ \end{aligned}$$

(18)

Appendix N: Values of alpha and KHI

$\alpha$	$o_{k,i}$	Description
0	0	Absence of sanctions
1	100	Sanction against an individual
10	50	Official announcement of sanctions
100	1	Sanction against a company
1,000	5	Sanction against an economic sector
3,000	15	Embargo

Because of their size, values of $\chi_{k,i,u}$ cannot be displayed as they are relative to the implementation date of each sanction (68 sanctions in total). We thus provide the value of $o_{k,i}$ as it defines the slope of ${\upchi }_{{{\text{k}},{\text{i}},{\text{u}}}}$. As stated in Bali and Rapelanoro (2021, p. 29): “The lower $o_{k,i}$ is, the more horizontal the slope will be, and the less the time factor will negatively impact the sanction’s ability to apply economic pressure […] In other words, $o_{k,i}$ is allowing us to calibrate the time factor intensity and behaviour.” The complete data base is available upon request at morad.bali@irnc.org.