Supplemental Appendix for “Does the Use of Imported Intermediates Increase Productivity? Plant-Level Evidence” (NOT FOR PUBLICATION)
Bibliographic record
Abstract
1.1 The issue of selection in the LP approach In this section we outline how we control for endogeneous selection while using LP intermediate proxy approach. The idea is essentially the same as the one used in Olley and Pakes (1996); namely, we first identify the state variables that are relevant for endogneous exiting decisions and approximate the survival probabilities using the polynomials in the observable variables. Then, we can control for the endogenous exiting decision by including the polinomials in the survival probabilities when the moment conditions are constructed. First, the state variables that are relevant for the plant exit decision are the predetermined level of capital kit and the past import decision di,t−1. The model in section 2 implies that a plant chooses to continue to produce if the current realization of productivity term ωit is higher than the threshold value ωt(kit, di,t−1). One might think that the intermediate proxy approach is not applicable to control for the selection bias because we cannot “recover ” ωit from observables given that we do not observe the current period intermediates if the plant chooses to exit. Note, however, that ωit follows the first order Markov process ωit = ξt + γdi,t−1 + ωi,t−1 + uit (equation (8) in the main text) and, thus, it is possible to approximate ωit using the observable variables (di,t−1, ωi,t−1), where ωi,t−1, in turn, can be proxied by the past value of intermediates, the past capital, and the past import decision so that ωi,t−1 = ω∗t−1(xi,t−1, ki,t−1, di,t−1) (equation (10) in the main text).
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.038 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.004 | 0.006 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.657 | 0.173 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".