1. Which geographic unit is larger? Commuting zones or Metropolitan Statistical Areas?

Ans: Metropolitan Statistical Areas (MSAs) are defined by the Census Bureau OMB (Office of Management and Budget) as an economic boundary different from commuting zones. MSAs include at least one county and only consider places with populations above 50,000.
There is no certain answer to this question because sometimes CZs > MSAs, sometimes MSAs > CZs.

2. How to define the suburbs in the first ring (mc=2), second ring (mc=3) ,or third ring (mc=4)? Why decompose the suburbs? Does it differ from aggregated one?

Ans. I mainly follow Census 1990 definition to collect MSA data.

  1.     Download the county shape file: https://www.census.gov/geographies/mapping-files/time-series/geo/carto-boundary-file.1990.html#list-tab-1556094155

2.     Insert the shape file into ArcGIS or QGIS, and duplicate the layer (if you want to create a map with two layers)

3.     “Layer” > “Open Attribute Table” (where you can see the data)

4.     “Layer” > “Filter”> Click the county name you want to keep  

5.     “View” > “Identify Features”

6.     Define which county are in first ring(mc=2): touch main county boundary(mc=1, where central city is located)

7.     Define which county are in second ring(mc=3): touch first-ring county boundary(mc=2) but not touch main county boundary(mc=1)

We want to see suburban local governments in different economic environments. Some suburbs are small in population, but some are large enough to be twin cities or satellite cities.

Aggregating all suburban local governments into one unit will average out the local characteristic (possible advantage: internalizing the coordination/competition within suburbs; 
possible disadvantage: might exist aggregation bias).

I mainly target top quartile MSAs (around 70 MSAs) in terms of the population in the Census 1990 definition, and I notice that most of them have first-ring suburbs. Few of them (around 10 MSAs) have first-ring and second-ring suburbs. Only MSAs in VA state have first-ring, second-ring, third-ring, and fourth-ring suburbs.

1. Which industry code system is the one you mainly follow? And why?

Ans: four-digit SIC industries; Pierce and Schott (2009) assign 10-digit HS products to four-digit SIC industries, ensuring each of the 397 manufacturing industries matches at least one trade code. 

2. What kinds of challengings do you face when you merge the data with different coding systems?

Ans: County business patterns (CBP) reports employment data by county
and industry for 6-digit NAICS codes in 2000. However, in 1980 and 1990, CBP reports by the industry for 4-digit SIC codes. Based on Census "bridge" file, we can construct a weighted crosswalk file. (Details upon request)

1. Why the Bartik instruments you construct is based on the 1980 employment share instead of 1986 or 1989 to proxy the 1990 local employment share?

Ans: County Business Pattern provides the most dis-aggregated employment data in Census years with more details than the remaining years.  

1. How to define trade shocks? What is the difference between Autor et al. (2013) and Autor et al. (2021)?

Ans: Generally, trade shocks are known as changes in import values per worker, import penetrations, or import exposure. However, Autor et al. (2021) define trade shocks as changes in import values divided by industry domestic absorption (US industry shipments plus net imports).

2. Why the evolving impact of the trade shock on local variables do not have consistent time differences in X and Y variables? To be specific, why is the form like: Y_{i,t+h}=a1+b1*Shock_{2000-2012}+X'_{it}*b2+e_{i,t+h}
   h=1,...12?

Ans: The baseline of the shock is from 2000 to 2012, one year before China joined WTO and after the culmination of the trade shock in 2010. Employees take time to migrate or find a non-manufacturing job, and we might only be able to see the complete adjustment along these margins after the shock reaches the peak. 

3. How about spillover effects between industries? Maybe China import shocks are not always generating adverse impacts on areas.

Ans. Acemoglu et al. (2016) collect US input-output data to build up supplier (upstream), and customer (downstream) import exposure shocks for both manufacturing and non-manufacturing industries. When customer industries are directly affected by trade shocks, the industries experience adverse employment effects. However, no evidence shows the employment changes in the industries when their suppliers are directly affected by trade shocks. 

1. Where do you obtain the government fiscal data? Do you have alternative sources?

Ans. I obtain government fiscal data from the Census Bureau: Annual Survey of State and Local Government Finances. The alternative source is: Pierson K., Hand M., and Thompson F. (2015). The Government Finance Database: A Common Resource for Quantitative Research in Public Financial Analysis. PLoS ONE doi: 10.1371/journal.pone.0130119
Most local governments usually report their fiscal data every five years, so I have to fill in the missing values (two methods: linear interpolation or constant growth rate).