Abstract
The fraction of US individual income tax returns filed as hand-prepared paper returns fell from 31 to 15% between 1999 and 2004, reflecting the spread of high-speed Internet and increased use of tax preparation software. I use zip code-level Internal Revenue Service tax data from 1998 to 2007 and Federal Communications Commission Internet service provider data to examine the effect of high-speed Internet availability on tax reporting behavior. Differences-in-differences results show increased rates of itemizing, higher total value of itemized deductions, and lower tax-to-income ratios after high-speed Internet becomes available. I differentiate between access in the tax year and the filing year and show that the tax reporting effects are driven by filing year availability and that filing year Internet availability also positively predicts electronic filing. I use instrumental variables methods to explore the relationship between Internet availability-induced electronic filing on tax reporting behavior and find suggestive evidence that e-filing driven by Internet access increased itemizing rates and total value itemized and lowered tax-to-income ratios. The results suggest changes in tax filing method driven by increased computer use and Internet access helped taxpayers identify relevant deductions and credits and lowered reported tax liability.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
H&R Block modified the ad slightly for the 2015 filing season, changing “billion” to “billions” and shifting the focus to the $304 billion in refunds that customers received in 2014 (Rooney 2015).
The Pew Internet and American Life Survey estimated that 34% of households had dialup connections and 3% had broadband connections. Estimates based on the 2000 Current Population Survey indicate that 42% of households had some type of Internet access in 2000.
This definition of high-speed Internet is now out of date. The FCC’s current broadband benchmark is 25 megabits per second for downloads and 3 megabits per second for uploads.
This figure is based on Fig. 1 in Dettling et al. (2018).
DGS iterate over coverage thresholds for urban zips (based on providers per thousand people) and for rural zips (based on providers per square mile), use the thresholds to construct aggregate coverage levels, and select the coverage thresholds that minimizes the root mean squared error between survey measures of Internet use and the threshold-based level. They select thresholds of one provider per 2700 people and one provider per 12 square miles as their coverage threshold. I replicate DGS’s measure with one exception: They adjust zip code population each year using the Statistics of Income zip code tax data filing rates as a measure of population change. I instead use a linear population trend between 2000 and 2010 Census data to interpolate population because tax filing rates for different populations change over time. See DGS’ (2018) “Appendix: Measuring broadband availability” for a detailed discussion of their alternative measure.
The zip code data are available for free from the IRS SOI division and can be downloaded from the IRS Web site at http://www.irs.gov/uac/SOI-Tax-Stats-Individual-Income-Tax-Statistics-zip-Code-Data-(SOI). Data for 1999, 2000, and 2003 are not publicly available.
Most returns are for the preceding tax year, but the data may include some late-filed returns or amended returns from prior years.
For panel specifications, I weight by the average number of returns in the zip code for 1998–2007.
The IRS no longer maintains active Web links to the e-file Demographic Data Web site on which it posted these data files. The files and documentation are available from the author or may be accessed through the Wayback Machine Internet archive at https://archive.org/web/. Files for 1998 are accessible through the Internet archive at http://www.irs.gov/prod/elec_svs/demogrfx.html. Files for 2000–2003 are available through the archive at http://www.irs.gov/efile/article/0,id=118376,00.html. Files for 2004–2010 are accessible through the archive at http://www.irs.gov/uac/e-File-Demographic-Data.
FCC data on the number of high-speed service providers by zip are available at https://transition.fcc.gov/wcb/iatd/comp.html.
Information on e-filing mandates is from http://www.taxadmin.org/fta/pub/mandates2.pdf and http://www.irs.gov/uac/The-following-states-have-mandated-e-file.
Results are similar when AGI is used instead of wage income.
In additional robustness checks (available from the author), I also instrument for e-filing using zip code-level 1990 housing density and percent multiple-family dwelling units (following Dettling (2017)). Locales with higher density or more MDUs lower the costs of providing high-speed Internet because the service provider can connect multiple households for each length of upgraded wiring. Kolko (2012) instruments for Internet access using the average slope of local terrain. I experiment with average slope as an instrument for Internet access, but find that the first-stage F-statistics in my slope specification are too weak to generate meaningful 2SLS results. The measures of slope generated by GIS software are sensitive to the shapefile used for zip code boundaries, the choice of elevation raster data, and the choice of map projection, which may explain differences between my first-stage results and those of Kolko (2012). Additionally, Kolko notes that the first-stage estimates are sensitive to the choice of weighting variables.
Table 3 shows results when standard errors are clustered at the commuting zone level and include zip and year fixed effects and commuting zone-specific time trends. Results are similar when standard errors are clustered at the state, county, or zip code level, with standard errors falling in magnitude as the size of the cluster shrinks. Point estimates are almost unchanged when commuting zone-specific time trends are excluded.
Appendix Table 1 shows these robustness tests for the “4 ISP” measure.
The 75th percentile corresponds to zip codes with more than 32% itemizers.
Measuring broadband availability as of June of the filing year and December of the tax year allows for the cleanest distinction between any high speed Internet availability in the tax year (which could affect labor supply) and Internet availability introduced after the end of the tax year. Results are similar when I measure tax year availability as of June of the tax year (available in Appendix Table 2), and weaker when I measure Internet availability as of December of the year preceding the tax year.
The value of itemized deductions was not included in the 2001 and 2002 tax data, so when 1998 is excluded the panel used for this outcome is 2004–2007, which is likely to explain the loss in significance.
There is no effect of tax year availability on average effective tax rates when using the 2001–2007 panel.
The event study results are very similar when the 2001–2007 sample is used instead, as shown in Appendix Fig. 1, although the tax year results are different. When using the “≥ 4 ISP” measure (see Appendix Fig. 2), the differences between measuring as of the filing year or the tax year are less apparent, and the results suggest a positive relationship between broadband availability and average effective tax rates.
These robustness test results are available in Appendix Table 4.
See Appendix Table 5 for these results, which are for regressions including controls for percent paid preparer returns and ln(average wage and salary income) and are weighted by number of returns per state. Results without these controls are similar and larger in magnitude.
The monetary value of EITC claims is not available for 2001.
References
Ashley, T., & Segal, M. A. (1997). Paid tax preparer determinants extended and Reexamined. Public Finance Review, 25(3), 267–284.
Atasoy, H. (2013). The effects of broadband internet expansion on labor market outcomes. Industrial and Labor Relations Review, 66(2), 315–345.
Borusyak, K., & Jaravel, X. (2017). Revising event study designs, with an application to the estimation of the marginal propensity to consume. Working paper.
Carter, L., Schaupp, L. C., & McBride, M. E. (2011). The U.S. e-file initiative: An investigation of the antecedents to adoption from the individual taxpayers’ perspective. e-Service Journal, 7(3), 2–19.
Chetty, R., Friedman, J., & Saez, E. (2013). Using differences in knowledge across neighborhoods to uncover the impacts of the EITC on earnings. American Economic Review, 103(7), 2683–2721.
Chetty, R., & Saez, E. (2013). Teaching the tax code: Earnings responses to an experiment with EITC recipients. American Economic Journal: Applied Economics, 5(1), 1–31.
Dettling, L. (2017). Broadband in the labor market: The impact of residential high-speed internet on married women’s labor force participation. Industrial and Labor Relations Review, 70(2), 451–482.
Dettling, L. J., Goodman, S. F., & Smith, J. (2018). Every little bit counts: The impact of high-speed internet on the transition to college. The Review of Economics and Statistics, 100(2), 260–273.
Dubin, J. A., Graetz, M. J., Udell, M. A., & Wilde, L. L. (1992). The demand for tax return preparation services. The Review of Economics and Statistics, 74(1), 75–82.
Duflo, E., Gale, W., Liebman, J., Orszag, P., & Saez, E. (2006). Savings incentives for low- and middle-income families: Evidence from a field experiment with H&R block. Quarterly Journal of Economics, 121(4), 1246–1311.
Erard, B. (1993). Taxation with representation. Journal of Public Economics, 52(2), 163–197.
Erard, B. (1997). Self-selection with measurement errors: A microeconometric analysis of the decision to seek tax assistance and its implications for tax compliance. Journal of Econometrics, 81(2), 319–356.
Faulhaber, G. R. (2002). Broadband deployment: Is policy in the way? In R. H. Crandall & J. H. Alleman (Eds.), Broadband: Should we regulate high-speed internet access (pp. 233–244). Washington, DC: AEI-Brookings Joint Center for Regulatory Studies.
Federal Communications Commission (FCC). (2009). High-speed services for internet access: Status as of December 31, 2007. Industry analysis and technology division, wireline competition bureau, January.
Federation of Tax Administrators. (2005). Electronic filing mandates: Lessons learned. Technical report, June.
File, T. (2013). Computer and internet use in the United States. Washington, DC: US Census Bureau.
Forman, C., Goldfarb, A., & Greenstein, S. (2012). The internet and local wages: A puzzle. American Economic Review, 102(1), 556–575.
Frecknall-Hughes, J., & Erich, K. (2015). Towards a general theory of tax practice. Social & Legal Studies, 24(2), 289–312.
Goolsbee, A. (2002). The turbo tax revolution? Evaluating the ability of technology to solve the tax complexity dilemma. In H. J. Aaron & J. Selemrod (Eds.), The crisis in tax administration (pp. 124–147). Washington, DC: Brookings Institution Press.
Government Accountability Office (GAO). (2009). Many taxpayers rely on tax software and IRS needs to assess associated risks. GAO-09-297, February 25.
Greenstein, S., & Prince, J. (2007). Internet diffusion and the geography of the digital divide in the United States. In R. Mansell, C. Avgerou, D. Quah, & R. Silverstone (Eds.), The Oxford handbook of information and communication technologies (pp. 168–195). Oxford: Oxford University Press.
Grunberg, D. B. (1998). Case study: Information technology at Jackson Hewitt Tax Service. Journal of Consumer Marketing, 15(3), 282–289.
Guldi, M., & Herbst, C. M. (2017). Offline effects of online connecting: The impact of broadband diffusion on teen fertility decisions. Journal of Population Economics, 30(10), 69–91.
Guyton, J. L., Korobow, A. K., Lee, P. S., & Toder, E. J. (2005). The effects of tax software and paid preparers on compliance costs. National Tax Journal, 58(3), 439–448.
Kolko, J. (2012). Broadband and local growth. Journal of Urban Economics, 71(1), 100–113.
Kopczuk, W., & Pop-Eleches, C. (2007). Electronic filing, tax preparers and participation in the Earned Income Tax Credit. Journal of Public Economics, 91(7), 1351–1367.
Kuhn, P., & Mansour, H. (2014). Is internet job search still ineffective? Economic Journal, 124(581), 1213–1233.
Kuhn, P., & Skuterad, M. (2004). Internet job search and unemployment durations. American Economic Review, 94(1), 218–232.
Leviner, S. (2012). The role tax preparers play in taxpayer compliance: An empirical investigation with policy implications. Buffalo Law Review, 60, 1079–1138.
Long, J. E., & Caudill, S. B. (1987). The usage and benefits of paid tax return preparation. National Tax Journal, 1993(40), 35–46.
Long, J. E., & Caudill, S. B. (1993). Tax rates and professional tax return preparation: Reexamination and new evidence. National Tax Journal, 46(4), 511–518.
Pew Research Center. (2017). Internet/broadband fact sheet, January 12. http://www.pewinternet.org/fact-sheet/internet-broadband/. Accessed 17 June 2018.
Rooney, J. (2015) H&R Block CMO Kathy Collins on the latest “billions” campaign and the quest for continued relevance. Forbes, January 9.
Treasury Inspector General for Tax Administration. (2006). Use of the free file program declined after income restrictions were applied. Reference Number 2006-40-171.
Treasury Inspector General for Tax Administration. (2013). Late legislation delayed the filing of tax returns and issuance of refunds for the 2013 filing season. Reference Number 2013-40-124.
Vigdor, J. L., Ladd, H. F., & Martinez, E. (2014). Scaling the digital divide: home computer technology and student achievement. Economic Inquiry, 52(3), 1103–1119.
Zelenak, L. A. (2010). Complex tax legislation in the turbotax era. Columbia Journal of Tax Law, 1, 91–119.
Zickuhr, K., & Smith, A. (2013). Home broadband 2013. Pew Research Center, August 26.
Acknowledgements
I gratefully acknowledge helpful comments from David Gunter, Kristin Hickman, Dan LaFave, Jim Siodla, Guillermo Vuletin, and two anonymous referees. Dan Meyer and Gary Koplik provided outstanding research assistance. All remaining errors are my own.
Author information
Authors and Affiliations
Corresponding author
Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Gunter, S.R. Your biggest refund, guaranteed? Internet access, tax filing method, and reported tax liability. Int Tax Public Finance 26, 536–570 (2019). https://doi.org/10.1007/s10797-018-9528-x
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10797-018-9528-x