Outline

1. Overview
   • Name and Theme of Dissertation (Proposal)
2. Chapter 1
   • Distance is the Soul of Beauty: How Open-Border Policy Realign Ideology?
3. Chapter 2
   • Anatomy of the China Syndrome: How China and WTO Impact Labor Welfare across Industries in Taiwan?
4. Chapter 3
   • The Calm before the Storm: A Cointegration Test of Political Risk and Trade Volume
5. To-do List
6. Timetable
7. Conclusion and Discussion

Overview of Dissertation (Proposal)

1. Title
   • Flouroscopy
     • Continuous X-rays Images; see FDA.gov
   • China Syndrome
     • Borrowed from trade literature
     • Autor, David H., David Dorn, and Gordon H. Hanson. 2013. “The China Syndrome: Local Labor Market Effects of Import Competition in the United States.” American Economic Review, 103 (6): 2121-68.
   • Front Line
     • Microscope China-Taiwan economic and political relationship
     • Use data published by Taiwanese governments
2. Theme
   • A dynamic analysis on China’s economic impact on Taiwan
   • Including: Trade Policy, Politics, Labor Market, and National Security

Chapter 1 | Is Distance the Soul of Beauty?

• Introduction
  • Old wisdom: Free Trade Agreement -> Economic Integration -> Political Integration
  • ECFA is a counter-example of this process
• Motivation
  • People tend to like each other when they are NOT familiar with each other.
  • This is also the case of China-Taiwan economic integration.

    • “…because of misunderstands unifies, because of understood that separates.”

• Main Questions
  • Question 1: “Do open border policies help the political integration?”
  • Question 2: “Does China-Taiwan case violate the progressive integration theory?”
  • Question 3: “Does the heterogeneous exposure to Chinese visitors realign the ideology?”
• Result Preview
  • Answer 1: No / Answer 2: Yes / Answer 3: Yes (The treatment effects in RD models become more significant after the shock)

Chapter 1 | Literature & Contribution

Chapter 1 | Trade Policy Shock

• In 2008,
  • Cross-strait Economic Cooperation Framework Agreement was rectified (ECFA Full Text)
  • Chinese Visitors to Taiwan dramatically increased

Chapter 1 | Ideology Realignment: Pre- v.s. Post - ECFA

• Fact
  • Bipartisanship. Blue: KMT, pro-China party; Green: DPP, against-China party
  • Pattern? Which regions switch their party recognition the most?
  • Main outcome variables in RD model is pro-China Margin (KMT vote - DPP vote)

Chapter 1 | Data Sources

Chapter 1 | Identification

• Methodology: Sharp Regression Discontinuity Design
  • cutoff (c): sample mean of the distances
• Identification
  • Sharp RD Design (Lee 2008; Lee and Lemieux 2010)

\[Y_{i} = D_i\tau + W_i \delta_1 + U_i \tag{RD1} \] \[D_i = 1[X_i \ge c] \tag{RD2} \] \[X_i = W_i\delta_2 + V_i \tag{RD3}\]

• Variables
  • \(Y_i\) : Outcome variables. K-D Margin, K%-D%
  • \(D_i\) : Treatment. The district is exposed to tourists or not.
  • \(X_i\) : Assigning variable. The travel time to TPE airport.
  • \(W_i\) : Unobserved endogenous variable. The real tourist numbers; which is assumed to have the effect on the result of election \(Y_i\).
  • According to tourist trade literature that distance negatively correlates with the number of tourists, so that \(\delta_2\) \(\neq 0\).

Chapter 1 | Result Preview: RD Graphs

• Different local polynomial assumptions

Chapter 1 | Algorithm

1. Calculate the distance between the municipalities and the tourists attractions.
2. Calculate the mean of distances, set the municipalities into control and treatment group.
3. Set the mean of the distance as the cutoff to the close and away regions.
4. Run the regression discontinuity using the 2004, 2008, 2012, 2016, 2020 presidential electoral data.
5. Compare the treatment effects across the different elections.
6. Regression those treatment effects on the importing number of the Chinese tourists. \[ \text{Treatment Effect}_t = \beta_0 + \Sigma_{s=0}^3 \beta_s \text{Chinese Tourist}_{t-s} + \varepsilon_t \]

Chapter 2 | Anatomy of Trade Openness and China Shock

Trade Openness Trend, source: DGBAS

Taiwan GDP Trend, source: World Bank

Chapter 2 | Main Question & Data

• How does trade volume affect the sector’s return of factors and profit?
  • Does China and joining WTO play a part?
  • Sectors: Agriculture, Manufacture, Service, and ICT
• Aggregate Level Data
  • Trade Volume = Export + Import
  • Total Profits of Firms
  • Total Return of Production Factors
• Trade Volume
  • Source
    • Directorate General of Budget, Accounting and, Statistics

Chapter 2 | Variables & Instruments

• Two main dependent variables

  • Total Profits of the firms
  • Total Returns of the production factors

• Benchmark OLS model \[\begin{align} \text{Total Profits}_\text{i,t} &= \nonumber \beta_0 + \beta_1 \text{Trade}_\text{i,t} +\beta_2\text{China}_\text{i,t} +\beta_3\text{WTO}_\text{i,t}+ \varepsilon_\text{t} \\ \text{Total Returns of Factors}_\text{i,t} &= \nonumber \beta_0 + \beta_1 \text{Trade}_\text{i,t} +\beta_2\text{China}_\text{i,t} +\beta_3\text{WTO}_\text{i,t}+ \varepsilon_\text{t} \end{align}\] \[\text{i} = \text{Agriculture(A), Manufacture(M), and Services (S)},\quad\text{t} = \text{Year 1981--2015}.\quad (\text{T}=35) \]

  • \(\text{China}_{it}\) and \(\text{WTO}_{it}\) are the binary dummy variables 0,1
  • I also explore the per capita level of data (divided by the total labor and firms)
  • IV and 2SLS results are the core of this chapter

• Instrument

  • CO2 permission per year in Taiwan
    • Source U.S. Energy Information Administration

Chapter 2 | 2SLS Results Summary

2SLS Model Result, Aggregate Data

• Note
  • WTO and China do NOT have a significant effect on both profits and returns
  • After adding the instrument, the Trade volume could explain both returns and profits.
  • Service sector surprisingly is the one affected the most by the trade.

Chapter 3 | Cointegration Test of Trade Volume and Political Stability

Main Question

• Is trade volume a bellweather of political stability?
  • Do the exporting or importing volume reflect the concurring political stability?
  • Do they have the long-run relationship?
  • Could we forecasting the political stability by using the trade data?
• Data Source
  • China-Taiwan industries trade value
    • Bereau of Foreign Trade
  • China-Taiwan political stability index
    • TheGlobalEconomy.Com

Chapter 3 | Step-by-setp Engle-Granger Cointegration Test

1. Download the trade volume between Taiwan and China in different industries (\(\text{trade}_{i,t}\))
2. Download the political stability index (\(\text{psi}_t\))
3. Run the unit root test, check the stationarity
4. Run the OLS model \(\text{trade}_{i,t} = \beta_0 + \beta_1 \text{psi}_t + \varepsilon_t\), esitmate the residual \(\hat{e}_t\)
5. Run the unit-root test on the residuals \(\hat{e}_t\) of the OLS model
6. Run the Error correction model, then check the Granger-Causality between \(\text{trade}_{i,t}\) and \(\text{psi}_t\)

Chapter 3 | Proposed VAR Model

• Economic Variables
  • Trade Volume (td), Monetary policy (i) , Exchange rate (xr), output (y), inflation rate (\(\Delta\)P)
• Non-Economic Variables
  • Political Stability Index (psi)
• Notations
  • \[ \text{Trade Volume} \rightarrow X_{1t}' = [td_t, y_t, xr_t, i_t, \Delta P_t]' \]
    
  • \[ \text{Political Stability Index} \rightarrow X_{2t}' = [psi_t, y_t, xr_t, i_t, \Delta P_t]' \]
• VAR model
  • stack \(X'_{1t}\) and \(X'_{2t}\) \[ X_t = \begin{bmatrix} X_{1t} & 0 \\ 0 & X_{2t} \end{bmatrix} \] \[ \Delta X_t = \Pi X_{t-1} + \Sigma^{k-1}_{i=1} \Gamma_i \Delta X_{t-1} + \mu_0 + \mu_i t + \varepsilon_t \]

Future work

1. Chapter 1

• Construct a trade theoretical model
• Explore the potential mechanism
• Add the other electoral results

2. Chapter 2

• Make IV and 2SLS become more convincing
• Run Unit-root Test and try a time-series approach
• Find out why the China Shock does not show up in the empirical model

3. Chapter 3

• Political stability index and Trade Dependencies on China.
• Run the Engle-Granger Cointegration Test.
• Build VECM or VAR model

Summary and Discussion