AdobeStock_271456045.jpeg
VoxEU Column International trade

Detecting the impact of policy uncertainty on foreign direct investment

Policy uncertainty is believed to affect foreign direct investment significantly, but, empirical evidence on the impact is scarce. This column proposes a unique empirical strategy for identifying the impact of a change in policy uncertainty on FDI at the sector level by utilising information about reservations of certain obligations contained in international investment agreements. It provides evidence that policy uncertainty discourages FDI. Policy uncertainty will continue to rise in the post-COVID-19 era, and the results highlight the importance of aligning globalisation with the mitigation of public health threats by implementing policies to reduce uncertainty.

Policy uncertainty – the uncertainty regarding future policies implemented by governments – has suddenly become a major impediment to globalisation. It affects firms' decisions related to significant sunk costs. An example of such a decision is foreign direct investment (FDI). Firms operating their business in foreign countries face uncertainty if the implementation of laws and regulations in their host countries can be reversed or undone. Recently, the spread of COVID-19 has further increased the level of uncertainty. Many countries are taking restrictive policy measures – the duration of which cannot be predicted – to prevent the spread of the virus, including limiting international trade and FDI (Baldwin and Freeman 2020). Consequently, in the post-COVID-19 era, the crucial policy issue is how to reconcile globalisation with the elimination of public health threats, which have the potential to cause high policy uncertainty.

In the trade literature, several studies have shown the important role of trade agreements, which facilitate trade by reducing trade policy uncertainty (Carballo et al. 2018, Handley and Limão 2015, 2017, Limão and Maggi 2015). The credibility of the government's commitment to lower tariff levels is enhanced by trade agreements and hence, trade policy uncertainty is reduced.1

Compared with the trade literature, studies examining how policy uncertainty affects FDI are surprisingly limited. Julio and Yook (2016) focus on FDI inflows from US firms to their foreign affiliates and examine the effects of policy uncertainty on FDI by using national election timing in host countries as a source of fluctuations in policy uncertainty. They find that US firms' FDI drops significantly during the period just before an election and increases after the uncertainty is resolved. Meanwhile, Azzimonti (2019) analyses the effect of partisan conflict in US trade policy on FDI inflows to the US. Using text from newspapers in the period 1985–2016, she constructs a new indicator to measure the degree of partisan conflict over trade policy and finds that an increase in partisan conflict over trade policy lowers FDI inflows to the US. 

Despite the importance of policy uncertainty for FDI, empirical studies in this field are still scarce. In a recent paper (Inada and Jinji 2020), we aim to contribute to the literature by investigating the impact of policy uncertainty on FDI through a unique approach based on information about international investment agreements (IIAs),2 which include bilateral investment treaties (BITs) and treaties with investment provisions. International investment agreements are “one of the few policy instruments that countries can use to directly attract foreign investment” (Egger and Merlo 2012: 1240).

Our main contribution is the proposal of a unique empirical strategy for identifying the impact of a change in policy uncertainty on FDI due to international investment agreements at the sector level. To the best of our knowledge, our work is the first of its kind in that it utilises the content of the negative lists in international investment agreements to capture the differences in policy uncertainty that foreign investing firms face in their host economies, depending on the sectors to which they belong. Our approach provides a new angle for exploring the sector-level relationship between policy uncertainty and FDI. Our method complements other approaches in existing empirical studies that mainly utilise national-level indexes, such as national election timing and partisan conflict.

The negative lists in an international investment agreement indicate which sectors are exempt from which obligations, such as national treatment (NT), most-favoured nation (MFN) treatment, and prohibition of performance requirement obligations. Unless sectors are included in the negative lists, FDI to those sectors is liberalised and protected. Thus, the signing of an international investment agreement with negative lists would leave the government (in particular, the government of a developing country with weak governance) little room for imposing discretionary restrictions on FDI for political reasons. This makes the exemptions plausibly exogenous to multinational enterprises (MNEs). The premise is that, after an international investment agreement enters into force, those sectors in a partner country would face different degrees of policy uncertainty from those of other sectors. In other words, after an international investment agreement enters into force, MNEs would face different degrees of policy uncertainty in the host country, depending on whether or not their sectors are exempt from certain obligations in the agreement; by contrast, they faced the same degree of policy uncertainty before the agreement enters into force.3

Two types of negative lists are available in the international investment agreement: (i) lists with standstill obligations ,and (ii) lists without standstill obligations (METI 2020). For lists with standstill obligations, “(1) measures that do not conform to the agreement cannot be newly introduced; (2) measures that do not conform to national treatment, most favoured nation, and performance requirement obligations that existed at the time the agreement became effective may be ‘maintained,’ but cannot be revised in a way that makes them more non-conforming to the agreement” (METI 2020: 614).4 For example, in the Japan–Laos IIA, the “manufacture of all types of motor vehicle” in Laos is exempted from national treatment obligation that are indicated in the list with standstill obligations. By contrast, the lists without standstill obligations are not subject to those obligations. For example, in the Japan–Chile IIA, “local basic telecommunication services and networks” in Chile are exempted from national treatment and most favoured nation obligations that are indicated in the list without standstill obligations. In our paper, we describe negative lists with standstill obligations as ‘current reservation lists’ and those lists without standstill obligations as ‘future reservation lists’.5

When an international investment agreement enters into force, sectors that are not included in the negative lists benefit fully from the agreement, whereas those included in the negative lists benefit only partially or not at all. Thus, the degree of policy uncertainty for those sectors decreases deterministically at the time of the agreement’s entry into force. On the one hand, although sectors included in the current reservation lists may not face more stringent regulations in the future, the firms in that sector do not know when, or in fact whether, regulations will be liberalised. Thus, in this sense, the degree of policy uncertainty for those sectors included in the negative lists decreases stochastically at the time of the agreement’s entry into force compared with those which are not included. Moreover, sectors included in the future reservation lists may face more stringent regulations in the future. Thus, the degree of policy uncertainty for those sectors remains unchanged or may even increase stochastically after an international investment agreement enters into force. Therefore, the degree of policy uncertainty will be higher in sectors included in the future reservation lists than those included in the current reservation lists.

Using these shocks for multinational enterprises, we can compare FDI in our treatment group (i.e. sectors that are included in negative lists) with FDI in our control group (i.e. sectors that are not included in negative lists) before and after an international investment agreement became effective.

We focus on 22 international investment agreements that were signed by Japan and entered into force by the end of 2016.6 Based on the content of the negative lists in each agreement regarding which obligations are reserved for which sectors, we implement difference-in-difference estimations to analyse the causal effects of policy uncertainty on FDI. We find evidence that policy uncertainty matters for Japanese outward FDI. In particular, we find that signing an international investment agreement stimulates FDI through a reduction in policy uncertainty, although the impact depends on the content of the agreement. Meanwhile, the impact may differ between national treatment and most-favoured nation exemptions. Whether sectors are subject to current reservation or future reservation may also matter. We find that the current reservation of most-favoured nation decreases capital investment. By contrast, the future reservation of national treatment increases the forecast error of capital investment. Finally, we find that the future reservation of most favoured nation reduces the share of the affected sector-country (economy) in the multinational enterprise’s global FDI activities in terms of affiliate investment.

As shown in Figures 1 and 2, the coefficients in the pre-treatment period are generally unchanged. By contrast, in the post-treatment period, although the changes are rather gradual, the coefficients tend to decline in capital investment (Figure 1) and to rise in forecast error of capital investment (Figure 2). This finding indicates that the trends in the treatment and control groups are likely to be parallel in the pre-treatment period but gradually divergent in the post-treatment period.

Figure 1 Estimated coefficients of most favoured nation exemptions on log capital investment

Note: The trend of log capital investment difference between affected sectors (treatment group) and no-change sectors (control group) is denoted by the dashed line. The shaded band represents the 95% confidence interval of the estimated effect.

Figure 2 Estimated coefficients of national treatment exemptions on log forecast error of capital investment

Note: The trend of forecast error of capital investment difference between affected sectors (treatment group) and no-change sectors (control group) is denoted by the dashed line. The shaded band represents the 95% confidence interval of the estimated effect.

The policy implications from our findings are straightforward. Policy uncertainty in host economies discourages globalisation. Thus, to maintain globalisation in the post-COVID-19 era, governments should implement policies which align global economic activities with the mitigation of public health threats.

Authors’ Note: The main research on which this column is based (Inada and Jinji 2020) first appeared as a Discussion Paper of the Research Institute of Economy, Trade and Industry (RIETI) of Japan.

References

Azzimonti, M (2019), “Does partisan conflict deter FDI inflows to the U.S.?”, Journal of International Economics 120: 162–178.

Baldwin, R and R Freeman (2020), “Supply chain contagion waves: Thinking ahead on manufacturing ‘contagion and reinfection’ from the COVID concussion”, VoxEU.org, 01 April.

Carballo, J, K Handley and N Limão (2018), “Economic and policy uncertainty: export dynamics and the value of agreements”, NBER Working Paper No. 24368.

Egger, P and V Merlo (2012), “BITs bite: an anatomy of the impact of bilateral investment treaties on multinational firms”, Scandinavian Journal of Economics 114(4): 1240–1266.

Inada, M and N Jinji (2020), “To what degree does policy uncertainty affect foreign direct investment? Micro-evidence from Japan's International Investment Agreements”, RIETI Discussion Paper Series 20-E-022, Research Institute of Economy, Trade and Industry.

Handley, K and N Limão (2015), “Trade and investment under policy uncertainty: theory and firm evidence”, American Economic Journal: Economic Policy 7(4): 189–222.

Handley, K and N Limão (2017), “Policy uncertainty, trade, and welfare: theory and evidence for China and the United States”, American Economic Review 107(9): 2731–2783.

Julio, B and Y Yook (2016), “Policy uncertainty, irreversibility, and cross-border flows of capital”, Journal of International Economics 103: 13–26.

Limão, N and G Maggi (2015), “Uncertainty and trade agreements”, American Economic Journal: Microeconomics 7(4): 1–42.

Ministry of Economy, Trade and Industry (METI) (2020), “The 2019 Report on Compliance by Major Trading Partners with Trade Agreements—WTO, EPA/FTA and IIA”, METI, Government of Japan.

United Nations Conference on Trade and Development (UNCTAD) (2006), Preserving Flexibility in IIAs: The Use of Reservations, UNCTAD, New York.

Tang, M and S Wei (2009), “The value of making commitments externally: Evidence from WTO accessions”, Journal of International Economics 78: 216–229.

Endnotes

1 For example, Handley and Limão (2015) find that Portugal's accession to the European Community (EC) removed uncertainty regarding future EC trade policy and helped increase Portuguese firms' export entry and sales. They use the gap between applied tariffs and tariff bindings as the measure of trade policy uncertainty because tariffs may be increased to the binding levels.

2 Egger and Merlo (2012) employ firm-level data of German multinational enterprises and their foreign affiliates in the period 1996–2005 and examine the effects of biletaral investment treaties. They find that both the signing and ratification of a bilateral investment treaty increase the numbers of investing firms per country, affiliates, employees, and investing sectors. However, their study does not address the issue of policy uncertainty on FDI through international investment agreements. In Inada and Jinji (2020), we aim to fill this gap in the research.

3 Reservation of obligations is not peculiar to IIAs signed by Japan. For example, it is included in the recent EU–Canada Comprehensive Economic and Trade Agreement (CETA). See International Investment Agreements Navigator by the United Nations Conference on Trade and Development here.

4 Moreover, once measures are revised to make them more consistent with the agreement, they cannot be made more inconsistent again (known as a ‘ratchet’ obligation to indicate that changes can only be made in one direction) (METI 2020, p. 614).

5 UNCTAD (2006) defines the former list as “reservations for existing measures and liberalization commitments” and the latter list as “reservations for future measures” (UNCTAD 2006, pp. 25–26).

6 It includes bilateral investment treaties and Economic Partnership Agreements with an investment chapter.

1,785 Reads