R Dataset / Package Ecdat / UStaxWords
Number of Words in US Tax Law
Thousands of words in US tax law for 1995 to 2015 in 10 year intervals. This includes income taxes and all taxes in the code itself (written by congress) and regulations (written by government administrators). For 2015 only "EntireTaxCodeAndRegs" is given; for other years, this number is broken down by income tax vs. other taxes and code vs. regulations.
number of words in thousands in the US income tax code
number of words in thousands in US tax code other than income tax
number of words in thousands in the US tax code
number of words in thousands in US income tax regulations
number of words in thousands in US tax regulations other than income tax
number of words in thousands in both the code and regulations for the US income tax
number of wrds in thousands in both code and regulations for US taxes apart from income taxes.
number of words in thousands in US tax code and regulations
Thousands of words in the US tax code and federal
tax regulations, 1955-2015. This is based on data
from the Tax Foundation (taxfoundation.org), adjusted
to eliminate an obvious questionable observation in
otherTaxRegulations for 1965. The numbers of
otherTaxRegulations was not reported
directly by the Tax Foundation but is easily computed
as the difference between their Income and Entire tax
numbers. This series shows the numbers falling by 48
percent between 1965 and 1975 and by 1.5 percent
between 1995 and 2005. These are the only declines
seen in these numbers and seem inconsistent with the
common concern (expressed e.g., in Moody, Warcholik
and Hodge, 2005) about the difficulties of simplifying
any governmental program, because vested interest
appear to defend almost anything. Lessig (2011)
notes that virtually all provisions of US law that
favor certain segments of society are set to expire
after a modest number of years. These sunset
provisions provide recurring opportunities for
incumbent politicians to extort campaign contributions
from those same segments to ensure the continuation of
the favorable treatment.
The decline of 48 percent in
otherTaxRegulations seems more curious for two
additional reasons: First, it was preceded by a
otherTaxRegulations between 1955
and 1965. Second, it was NOT accompanied by any
comparable behavior of
the latter grew each decade by between 17 and 53
percent, similar to but slower than the growth in
otherTaxRegulations for 1965 is
replaced by the average of the numbers for 1955 and
EntireTaxRegulations for 1965 is
comparably adjusted. This replaces (1322, 2960) for
those two variables for 1965 with (565, 2203). In
EntireTaxCodeAndRegulations are also changed
from (1626, 3507) to (870, 2751).
Independent of whether this adjustment is correct or not, it's clear that there have been roughly 3 words of regulations for each word in the tax code. Most of these are income tax regulations, which have recently contained 4.5 words for every word in code. The income tax code currently includes roughly 50 percent more words than other tax code.
Tax Foundation: Number of Words in Internal Revenue Code and Federal Tax Regulations, 1955-2005 Scott Greenberg, "Federal Tax Laws and Regulations are Now Over 10 Million Words Long", October 08, 2015
J. Scott Moody, Wendy P. Warcholik, and Scott A. Hodge (2005) "The Rising Cost of Complying with the Federal Income Tax", The Tax Foundation Special Report No. 138.
data(UStaxWords) plot(EntireTaxCodeAndRegs/1000 ~ year, UStaxWords, type='b', ylab='Millions of words in US tax code & regs')# Write to a file for Wikimedia Commons ## Not run: svg('UStaxWords.svg')## End(Not run) matplot(UStaxWords$year, UStaxWords[c(2:3, 5:6)]/1000, type='b', bty='n', ylab='', ylim=c(0, max(UStaxWords$EntireTaxCodeAndRegs)/1000), las=1, xlab="", cex.axis=2) lines(EntireTaxCodeAndRegs/1000~year, UStaxWords, lwd=2) ## Not run: dev.off()## End(Not run) # lines 1:4 = IncomeTaxCode, otherTaxCode, # IncomeTaxRegulations, # and otherTaxRegulations, respectively## ## Plotting the original numbers without the adjustment ## UStax. <- UStaxWords UStax.[2,c(6:7, 9:10)] <- c(1322, 2960, 1626, 3507) matplot(UStax.$year, UStax.[c(2:3, 5:6)]/1000, type='b', bty='n', ylab='', ylim=c(0, max(UStax.$EntireTaxCodeAndRegs)/1000), las=1, xlab="", cex.axis=2) lines(EntireTaxCodeAndRegs/1000~year, UStax., lwd=2) # Note especially the anomalous behaviour of line 4 = # otherTaxRegulations. As noted with "details" above, # otherTaxRegulations could have tripled between 1955 # and 1965, then fallen by 48 percent between 1965 and # 1975. However, that does not seem credible, # especially since there was no corresponding behavior # in otherTaxCode.## ## linear trend ## (newWdsPerYr <- lm(EntireTaxCodeAndRegs~year, UStaxWords)) plot(UStaxWords$year, resid(newWdsPerYr)) # Roughly 150,000 additional words added each year # since 1955. # No indication of nonlinearity.
Dataset imported from https://www.r-project.org.
From Around the Site...
|Title||Authored on||Content type|
|npk||February 26, 2017 - 11:28 AM||Dataset|
|R Dataset / Package HSAUR / agefat||March 9, 2018 - 1:06 PM||Dataset|
|R Dataset / Package robustbase / epilepsy||March 9, 2018 - 1:06 PM||Dataset|
|faithful||February 26, 2017 - 11:28 AM||Dataset|
|R Dataset / Package DAAG / grog||March 9, 2018 - 1:06 PM||Dataset|