Estimating the trade and welfare effects of Brexit: A panel data structural gravity model
نویسندگان
چکیده
This paper proposes a new panel data structural gravity approach for estimating the trade and welfare effects of Brexit. Assuming different counterfactual post-Brexit scenarios, our main findings suggest that UK's exports goods to EU are likely decline within range between 7.2% 45.7% six years after Brexit has taken place. For UK, negative only partially offset by an increase in domestic with third countries, inducing real income 0.3% 5.7%. The estimated not from zero, but some members like Ireland expected also experience losses. Évaluation des effets du sur le commerce et bien-être à l’aide d’un modèle de gravité structurel s’appuyant données longitudinales. Afin d’évaluer les commerce, nous proposons dans cet article un nouveau En supposant plusieurs scénarios contrefactuels, nos principaux résultats suggèrent que ans après Brexit, exportations biens Royaume-Uni vers l’Union européenne pourraient chuter volumes compris entre 7,2 % 45,7 %. Pour Royaume-Uni, négatifs ne seront partiellement compensés par une augmentation intérieur échanges avec pays tiers, ce qui engendrera diminution revenu réel comprise 0,3 5,7 Au sein européenne, relativement au bien-être, conséquences nulles, mais certains membres comme l’Irlande subir pertes. On Thursday, June 26, 2016, United Kingdom held “Brexit referendum,” majority participating electorate voting favour “leave choice.” As consequence, David Cameron resigned as prime minister Theresa May took office. March 29, 2017, government UK officially handed letter Brussels notifying country's withdrawal European Union, triggering Article 50 Treaty on Union. initiated two-year window conclude agreement. Since then, have been negotiating terms withdrawal. December 14, 2018, both parties announced conclusion agreement began working political declaration outlined issues should govern future bilateral relationships. has, since this time, failed convince parliamentary and, none politically discussed potential outcomes can be ruled out end February 2019. During period campaigning prior (and after) referendum, economic costs benefits induced highly debated. Given fact negotiators still far reaching final all involved issues, (ex ante) estimation is naturally surrounded substantial policy uncertainty. Nevertheless, studies tried provide estimates focusing issues. In these studies, areas attracted most attention among economists policy-makers alike. cases, models quantitative literature extended customized order calculate stemming With one exception, available analyses point towards (maybe substantial) reduction remaining 27 member states consequence interaction would accompanied areas. magnitude differs depending approaches applied, used importantly, scenarios assumed.11 A more detailed account other applying (structural) offered section 2. question Brexit-induced specific application evaluation. Policy-makers public at large relatively high demand such exercises, ex ante post. Other recent initiatives broad interest Transatlantic Trade Investment Partnership (TTIP) Comprehensive Economic Agreement (CETA). reason, it seems essential framework assessing measures consistent standard theory accounts uncertainty stems direct partial effects. provides additional information calculation meaningful bandwidths applies novel identifying unifying spirit Allen al. (2020) takes example. particular, we extend constrained Poisson pseudo-maximum likelihood estimator (CPPMLE), suggested Pfaffermayr (2020), full endowment general equilibrium effects, Yotov (2016). CPPMLE features advantageous properties useful First, avoids set dummy variable parameters exploiting restrictions imposed system multilateral resistances, purposes (see French 2016 online technical appendix details). effect, treats bilateral, exporter–time importer–time dummies functions time-varying friction indicators observed data. Thus, able accurately address solve incidental parameter problem22 problem arises because small number observations estimate fixed while consistently. However, even if eliminates (by transformation), sample size. Concise econometric results issue yet available. allows fully control unobserved heterogeneity across country pairs. important property (follow-up) analysis. Second, delivers unbiased errors slope reliable inference. contrast, demonstrated (2019), (PPML) many suffers downward-biased errors. reason lies leverage found increasing Third, apply delta method33 method designed confidence intervals non-linear prediction flows pairs under 2018 general, intuitive technique approximating moments random variables (e.g., conditional variance) making use polynomial truncated Taylor series centred mean (Oehlert 1992). calculating theory-consistent intervals, which when various alternative scenarios. Monte Carlo simulations provided (2018) indicate predictions do suffer finite bias, coverage rates correct panels minimum about 60 economies. might preferable compared routinely applied bootstrapping procedures. context models, resistance together computationally intensive, theoretical reliability obtained empirical specifications model allow phasing-in free agreements UK. purpose, follow Bergstrand (2015) distributed lag structure, considering contemporaneous identify lower bound estimates. Further, time trends border investigates sensitivity respect identification associated measures. previous literature, membership either identified means average effect existing regional (RTAs) or accounting indicators. former measure too its definition RTAs substantially differ their respective depths concerning degree liberalization. datasets capturing periods, seem driven eastern enlargement, obtain good (additional) alternative, suggests using customs unions (direct) empirically compares specification commonly involving (only) RTA indicators, Baier (2018). Most dealing relies versions periods World Input–Output Database (WIOD) (see, e.g., Brakman Dhingra Felbermayr 2017 Vandenbussche 2017). WIOD contains mainly developed economies around world, very actively engaged policies. source lacks exploitable (time-)variation makes difficult causal and/or unions. Furthermore, input–output tables technically constructed way sum residuals relationships zero. representation constitutes drawback statistical inference puts error any applied. reasons, rely unique dataset combines sources trade, total production manufacturing goods.44 detail A1. four These based two generate necessary subsequent analysis first (FTAs) flows. aims trade-creating custom versus FTAs. second uses single indicator presence (RTA) common thus insights into choice resulting regard negotiations, distinguish “hard” “soft” hard scenario assumes will leave lose current countries. further no could established. countries world Organization regulations. data, trading preferential force. specifications. one, loses union relationship EU-27 gains FTAs specification, simultaneously zero part. Using newly established FTA soft scenario. closely mimics “Global Britain” strategy proposed recently official position.55 See, example, www.theguardian.com/politics/2018/jul/07/brexit-summit-how-the-papers-saw-theresa-mays-deal. scenario, net given difference inherits remain does FTAs, state All again Our reveal following findings. distorting over pointing importance cumulative trade-enhancing qualitatively line (2015), somewhat smaller. explained shorter span covered lasting 1994 2012. differentiation important, increases significantly larger amount latter, Six formation increased approximately 64%. corresponding amounts 33%. Pooling those policies simple deliver accurate EU-27, respectively. much 39%, closer specification. conducted reduces EU–UK pronounced imports EU. Not surprisingly, largest differentiating impact quantitatively largest, finding surprising. predicts decrease (EU) (UK) 35.5% (29.4%). addition, Under varies 25.3% 45.7%. exhibit “positive” diversion identifies positive intra-EU reveals cut than half. reduced 17% (14%) case ranges large, indicated calculated intervals. When model, was smaller union. “negative” creation via overall effects.66 changes Costinot Rodríguez-Clare (2014) interpreted income. Across specifications, that, (statistically significant) (under Brexit) 5.7% Brexit), EU, statistically never points asymmetry borne remainder structured follows. Section 2 summarizes methodological comparability, limit discussion contributions estimation. 3 presents 4 discusses details CPPMLE, reports model. 5 assumed. 6, offer concluding remarks discuss implications. scheduled referendum status triggered aimed win exit campaign. potentially relevant dimensions investment productivity Baldwin (2016), Van Reenen (2016) Sampson (2017). following, concentrates scientific put centre investigations studying Table 1 brief overview reviewed displaying utilized reported. vote national international (governmental) institutions, including HM Treasury Kierzenkowski International Monetary Fund Gudgin ad hoc formulations take type explicitly account. report prepared aggregated flow (in logs) Glick Rose database covering 200 1948 2013 (mainly) estimators gravity-type modelled due market (European Area) export included isolating membership. Applying (atheoretical) assuming WTO rules aftermath reductions amounting −20% case. signing add up −17%. deviate slightly modelling captured but, terms, similar (2017) comprehensively investigate robustness above mentioned institutions by: (i) mimicking (ii) varying composition, considered authors alternatively run relying estimator, biased inherently heteroscedasticity (Santos Silva Tenreyro 2006). checks upper carried understanding short-run long-run research team London School Economics Political Science Centre Performance (Dhingra simulate calibrates value added year 2011, 40 35 industrial sectors. Accordingly, losses smallest, −1.3%, market. rules, households doubled −2.7%. form regressions, furthermore document severe setbacks competitiveness) foreign investments (FDI). tripled spread almost evenly whole distribution. develop sector-level linkages separately study employment each state. inferred sectoral elasticities Imbs Méjean differentiates imply non-tariff barriers (NTBs). defined Area NTBs table scenarios). 2014 1.21% (soft 4.47% (hard job 140,000 530,000 jobs. absolute jobs lost ranging 280,000 1.2 million. related work paper. taking equation PPML iterative procedure possible counterfactually change (MLR) level (i.e., effects). counterfactuals agreements. trades retains assume partners Empirically, cross-sectional 43 Their drop 18% 14% strategy, signs non-EU database, sufficient vein, paper, however, exploits variation spanning 2000 2014, allowing heterogeneous providing levels disaggregation, Similar (2017), enter inputs multi-country, multi-sector Ifo Institute Research. paper's regulations induce loss 2%. Within asymmetrically distributed, services negatively affected cases reconsiders framework, constraints resistances. specifically, following. comparative static analysis, assess bandwidth formulate conclusions. ignores
منابع مشابه
A Panel Data Analysis of South Korea’s Trade with OPEC Member Countries: The Gravity Model Approach
Abstract T his paper explains bilateral trade patterns between South Korea and thirteen OPEC member countries over the period 1980-2014 using a gravity model. The estimation results show that the gravity equation fits the data reasonably well. We confirmed the existence of long term relationships between the bilateral trade flows and the main components of gravity model - ...
متن کاملa study on insurer solvency by panel data model: the case of iranian insurance market
the aim of this thesis is an approach for assessing insurer’s solvency for iranian insurance companies. we use of economic data with both time series and cross-sectional variation, thus by using the panel data model will survey the insurer solvency.
a panel data analysis of south korea’s trade with opec member countries: the gravity model approach
abstract t his paper explains bilateral trade patterns between south korea and thirteen opec member countries over the period 1980-2014 using a gravity model. the estimation results show that the gravity equation fits the data reasonably well. we confirmed the existence of long term relationships between the bilateral trade flows and the main components of gravity model - gdp, income (gdp per c...
متن کاملThe Gravity Model and Iran's Trade Flows
This article has considered the volume and direction of Iran’s trade using the gravity model. The major issue in this analysis is to explore why Iran over or under-traded with the 76 countries relative to the predicted trade flows of the model. The study attempts to explore the reasons from the respect of both the model itself and Iran’s trade structure. This is done by analysing the performanc...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Canadian Journal of Economics
سال: 2021
ISSN: ['0008-4085', '1540-5982']
DOI: https://doi.org/10.1111/caje.12494