From dc93eec051408a4e5b103974b18b852315d393be Mon Sep 17 00:00:00 2001 From: Claude Date: Wed, 4 Nov 2026 20:12:19 +0000 Subject: [PATCH] docs: check the survey-year rule against the 2009 stimulus (#77) A second natural experiment for the state-government rule in R/fiscal.R, using IG Federal revenue from the public API (corpus build 2026-10-26, fccd87c), survey years 2006-2011: - Alabama and Michigan (September 30) and Texas (August 31) jump in survey year 2009 (+18.0%, +22.2%, +12.3%) and slow in 2011 (-4.7%, +3.9%, +4.3%), as the June 30 states do (median +12.4%, +3.6%). Under the plain rule their 2009 would end in 2008, before the stimulus, and their 2011 would be the full stimulus year. - 15 large September-30 cities and counties jump in 2011 (median +17.7%, against +2.1% and +3.1% in 2009 and 2010), as the plain rule predicts. - D.C. is inconclusive again: its 2009 is flat (+4.1%), like the plain rule, but its 2011 falls (-0.9%), like the documented one. The data agrees with the code, so the rule and .SURVEY_YEAR_SAME_YEAR_GOVIDS are unchanged. The header comment and the "survey-year rule" section now cite survey years 2009 and 2020 for the state rule and say D.C. was inconclusive in both. man/ regenerated with roxygen2 7.3.3. Refs #77 Co-Authored-By: Claude Opus 5.5 Claude-Session: https://claude.ai/code/session_01Qbe6u2h83wgZdTgJgE4Xh5 --- R/fiscal.R | 33 +++++++++++++++++++++------------ man/cog_fiscal_periods.Rd | 12 +++++++----- man/cog_survey_year.Rd | 12 +++++++----- 3 files changed, 35 insertions(+), 22 deletions(-) diff --git a/R/fiscal.R b/R/fiscal.R index 31e154a..da14874 100644 --- a/R/fiscal.R +++ b/R/fiscal.R @@ -13,14 +13,21 @@ # * State governments (type 0). Census's 2004 and 2006 technical # documentation names the Alabama and Michigan (September 30) and Texas # (August 31) state governments as exceptions "included as though they -# were part of the June 30 group". The published corpus agrees for survey -# year 2020: those three states' federal aid jumps with the CARES Act in -# 2020, as the June-30 states' does, while New York (March 31) and 15 -# large September-30 cities and counties jump in 2021 (measurements in -# uscogdata#77). For the other 47 states both rules give the same months. +# were part of the June 30 group". The published corpus agrees in two +# survey years (measurements in uscogdata#77): +# - 2020: those three states' federal aid jumps with the CARES Act in +# 2020, as the June-30 states' does, while New York (March 31) and 15 +# large September-30 cities and counties jump in 2021. +# - 2009: their federal aid jumps with the 2009 stimulus in survey year +# 2009 and slows in 2011, as the June-30 states' does. Under the plain +# rule their 2009 would end in 2008, before the stimulus, and their +# 2011 would be the full stimulus year. The same 15 local governments +# jump in 2011, as the plain rule predicts for them. +# For the other 47 states both rules give the same months. # * The District of Columbia (type 2 here), which the same documentation -# lists with the September-30 exceptions. The 2020 data neither confirms -# nor rules this out, so it rests on the documentation alone. +# lists with the September-30 exceptions. The 2009 and 2020 data each +# neither confirm nor rule this out, so it rests on the documentation +# alone. # canonical_govids that follow the state-government rule without being type 0. .SURVEY_YEAR_SAME_YEAR_GOVIDS <- c( @@ -218,11 +225,13 @@ #' 30) that means October 2016 to September 2017, and for Texas (August 31) #' September 2016 to August 2017. Census's 2004 and 2006 technical #' documentation lists these three states as exceptions to the June 30 window, -#' and the corpus agrees in survey year 2020, the one year checked. The same -#' documentation lists the District of Columbia, so it is treated the same -#' way. That treatment has not been checked against the data. For New York -#' (March 31) and the 46 states with a June 30 year end, the two rules give -#' the same months. +#' and the corpus agrees in survey years 2009 and 2020, the two years checked: +#' their federal aid moves with the 2009 stimulus and the 2020 CARES Act in +#' the same survey years as the June 30 states' does. The same documentation +#' lists the District of Columbia, so it is treated the same way. Checks +#' against the 2009 and 2020 data were inconclusive for it, so that treatment +#' rests on the documentation alone. For New York (March 31) and the 46 states +#' with a June 30 year end, the two rules give the same months. #' #' A government can change its fiscal year end, so the periods are looked up #' per government-year. Where the corpus has no `fiscal_year_end` for a diff --git a/man/cog_fiscal_periods.Rd b/man/cog_fiscal_periods.Rd index 5b5448f..9db33bc 100644 --- a/man/cog_fiscal_periods.Rd +++ b/man/cog_fiscal_periods.Rd @@ -48,11 +48,13 @@ fiscal year ending in calendar year Y. For Alabama and Michigan (September 30) that means October 2016 to September 2017, and for Texas (August 31) September 2016 to August 2017. Census's 2004 and 2006 technical documentation lists these three states as exceptions to the June 30 window, -and the corpus agrees in survey year 2020, the one year checked. The same -documentation lists the District of Columbia, so it is treated the same -way. That treatment has not been checked against the data. For New York -(March 31) and the 46 states with a June 30 year end, the two rules give -the same months. +and the corpus agrees in survey years 2009 and 2020, the two years checked: +their federal aid moves with the 2009 stimulus and the 2020 CARES Act in +the same survey years as the June 30 states' does. The same documentation +lists the District of Columbia, so it is treated the same way. Checks +against the 2009 and 2020 data were inconclusive for it, so that treatment +rests on the documentation alone. For New York (March 31) and the 46 states +with a June 30 year end, the two rules give the same months. A government can change its fiscal year end, so the periods are looked up per government-year. Where the corpus has no `fiscal_year_end` for a diff --git a/man/cog_survey_year.Rd b/man/cog_survey_year.Rd index ef9d19a..fdaeeed 100644 --- a/man/cog_survey_year.Rd +++ b/man/cog_survey_year.Rd @@ -59,11 +59,13 @@ fiscal year ending in calendar year Y. For Alabama and Michigan (September 30) that means October 2016 to September 2017, and for Texas (August 31) September 2016 to August 2017. Census's 2004 and 2006 technical documentation lists these three states as exceptions to the June 30 window, -and the corpus agrees in survey year 2020, the one year checked. The same -documentation lists the District of Columbia, so it is treated the same -way. That treatment has not been checked against the data. For New York -(March 31) and the 46 states with a June 30 year end, the two rules give -the same months. +and the corpus agrees in survey years 2009 and 2020, the two years checked: +their federal aid moves with the 2009 stimulus and the 2020 CARES Act in +the same survey years as the June 30 states' does. The same documentation +lists the District of Columbia, so it is treated the same way. Checks +against the 2009 and 2020 data were inconclusive for it, so that treatment +rests on the documentation alone. For New York (March 31) and the 46 states +with a June 30 year end, the two rules give the same months. A government can change its fiscal year end, so the periods are looked up per government-year. Where the corpus has no `fiscal_year_end` for a