PR for #268 and more: revise updated manuscript

[santiagohermo]

Can you take a look at the paper @diegogentilepassaro @gabrieleborg? For comparison, the version as of the AER submission can be found here.

I will review the paper myself, make a few corrections, and submit on Monday. It'd be great to have your feedback, but no problem if you don't have time.

What's changed relative to previous draft

• Title and title page (see thank you note)
  • Should we use "From Workplace to Homeplace" instead of "From Workplace to Residence"? It sounded good to me but I don't remember why we discarded it.
• Introduction
  • Changed tone to focus on estimates of workplace MW, mentioning that no hard conclusions can be drawn from negative coefficient on residence MW
  • Highlight that the fact that the Zillow data reflects posted rents in a month is positive
  • Update limitations
• Model
  • Add section 2.3, where we discuss the model based on the comments we got
  • Extend theory appendix to include model with endogenous commuting shares
• Context and Data
  • Add Online Appendix Figure 2
  • Discuss posted rents vs contract rents
  • Introduce ZORI index
• Empirical strategy
  • Revise write-up, putting the identification discussion in Section 4.3
  • Focus on residence MW as key control
  • Add worry of changing composition of listings and how ZORI helps with that
• Empirical results
  • Again, focus on workplace MW. Figure 4 only adds leads and lags of workplace MW, and shows implied cumulative effects (I tried estimating them, but they don't look significant if we cluster at state level)
  • Add leads and lags of residence MW in appendix
  • Add discussion of Zillow's repeat rental index
  • Update discussion of state-time FE model, saying it's consistent with baseline
  • There was an error when standardizing variables for the heterogeneity table: now it's fixed
  • Change variables used to compute weights for reweighted model. Drop "unbalanced reweighted" column
  • Drop "Other Geographies and time frames" and related table, keeping only dynamic county-level model
  • Extend "Summary and Discussion" section, discussing estimates of income elasticity to the MW. Does this review make sense to you @gabrieleborg?
• Counterfactual
  • Use elasticity of income to the MW of 0.1 as baseline, show estimates changing the elasticity
  • Some improvements in writing

[gabrieleborg]

That's amazing @santiagohermo ! I will start the review now. I will finish post comments during the weekends.

[santiagohermo]

Thanks @gabrieleborg! Feel free to focus on the changes relative to the previous version. You may also look at the code if you have doubts.

[gabrieleborg]

Finished reviewing the paper. Amazing job @santiagohermo !! here's the PDF with my comments, and suggested edits. When I provide no context in the comment, you should simply take that as a suggested edit. Feel free to accept it or not :)

As for the extended "Summary and Discussion" session, I like it and I think it's clear. I didn't have comments on that.

min_wage_rent (1).pdf

[santiagohermo]

Thanks @gabrieleborg! I finished my review myself, I'll proceed to implement changes and get ready for a new submission

[santiagohermo]

Here are some replies to your comments @gabrieleborg: min_wage_rent.gb.sh.pdf

I implemented many (see https://github.com/diegogentilepassaro/min_wage_rent/pull/271/commits/23a70a7bb1d7d901f274bb427f06b9de62eb1d41). Of the ones I didn't implement, I provided explanations for some. I mostly ignored a few that were about things that did not change relative to our previous draft (so were already reviewed extensively)

[santiagohermo]

Summary here