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 words in
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
The decline of 48 percent in
otherTaxRegulations seems more
curious for two additional reasons: First,
it was preceded by a tripling of
otherTaxRegulations between 1955 and
1965. Second, it was NOT accompanied by
any comparable behavior of
otherTaxCode. Instead, the latter
grew each decade by between 17 and 53
percent, similar to but slower than the
1965 is replaced by the average of the
numbers for 1955 and 1975, and
EntireTaxRegulations for 1965 is
comparably adjusted. This replaces (1322,
2960) for those two variables for 1965
with (565, 2203). In addition,
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.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70
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$year, resid(newWdsPerYr)) # Roughly 150,000 additional words added each year # since 1955. # No indication of nonlinearity. # adusted R-squared exceeds 99 percent. ## ## linear trend with increased slope ## during the Reagan years ## # linear spline with knots at # 1981 and 1989 Reagan <- pmax(0, pmin( (UStaxWords$year-1981)/8, 1)) plot(Reagan~year, UStaxWords, type='b') UStaxWords$Reagan <- Reagan ReaganMdl <- EntireTaxCodeAndRegs~year + Reagan fitReagan <- lm(ReaganMdl, UStaxWords ) summary(fitReagan)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.