São Paulo’s PIPE Grants: Dose and Innovation

Overview

This series uses São Paulo’s PIPE program to ask whether larger R&D grants reliably produce more innovation in small and medium-sized firms. It moves from the program’s applicant and measurement lifecycle, through the same-call counterfactual and delayed patent outcome, to a dose–response curve, its plausible absorptive-capacity explanation, robustness checks, and policy calibration. The study finds more patenting for funded applicants on average and an inverted-U relationship in the observed grant distribution, with the modeled peak at 59% of its normalized intensity scale. That result concerns fractional patent applications over defined horizons, not all innovation or social value; the mechanism is plausible rather than directly identified, and the peak is specific to PIPE’s applicants, rules, outcomes, and data.

Learning outcomes

Within PIPE’s applicant population, funded firms registered more fractional patent applications than matched non-funded applicants over a delayed outcome window.
The modeled dose–response is inverted U-shaped: marginal patent gains rise through an intermediate range and decline at higher doses.
The 59% intensity peak and approximate US$340,655.18 conversion describe PIPE’s observed distribution, not a universal funding target.
Absorptive capacity is a plausible explanation for the curve, not a directly identified mediation pathway in the study.
Patent results do not establish equivalent effects on sales, productivity, patent quality, spillovers, or total social value.

Keywords

PIPEInnovationGrantsPatentsDose–responseAbsorptive capacityCounterfactualSMEsFAPESPSão Paulo

Chapters

PIPE Grant Dose and Patenting

Why receipt alone misses project fit and capacity

10 min · Why receipt alone misses project fit and capacity

10 min

PIPE Applicants and Patent Data

How program records create the comparison frame

8 min · How program records create the comparison frame

8 min

PIPE Counterfactual and Patent Effects

How same-call applicants sharpen causal inference

9 min · How same-call applicants sharpen causal inference

9 min

PIPE Dose–Response Curve

What the normalized intensity scale actually means

9 min · What the normalized intensity scale actually means

9 min

PIPE Absorptive Capacity and Robustness

What robustness supports—and leaves unresolved

9 min · What robustness supports—and leaves unresolved

9 min

PIPE Grant Policy Calibration

How to match support to projects and firms

9 min · How to match support to projects and firms

9 min

PIPE Grants: When More Helps

A conditional answer about dose, capacity, and context

9 min · A conditional answer about dose, capacity, and context

9 min

Preview every chapter · Unlock full text, audio, and video from chapter 4

Plans

Ch.1-3 FreeCh.4-7 Premium

FAQ

How credible is the PIPE grant comparison?

It is more credible than a raw recipient-versus-nonrecipient comparison because it uses same-call applicants, pre-treatment adjustments, fixed effects, and delayed outcomes. It remains observational and depends on assumptions about unmeasured selection and grant intensity.

Did PIPE grants increase patenting?

For the studied São Paulo applicant population, PIPE funding was associated with an average increase of 0.155 fractional patents relative to matched non-funded applicants over three years. That estimate concerns patenting, not every form of innovation or social value.

Does a larger PIPE grant always produce more innovation?

No. The modeled patent response rises through an intermediate observed dose and declines at higher doses. A lower marginal patent response does not prove that a large grant has no later, unmeasured, or broader value.

Can PIPE’s 59% dose be copied as a funding rule?

No. The 59% point is a modeled maximum on PIPE’s normalized intensity scale and depends on its applicants, rules, grant distribution, outcome, horizon, and model. It can inform local analysis, not set a universal award.