Abstract
Recent studies of shadow economies focus primarily on cross-country comparisons. Few have examined regional or state-level variations in underground economic activity. This paper presents estimates of the shadow economy for 50 US states over the period 1997–2008. Results suggest that tax and social welfare burdens, labor market regulations, and intensity of regulation enforcement are important determinants of the underground economy. Among the states, Delaware, on average, maintains the smallest shadow economy at 7.28 % of GDP; Oregon, on average, has the second smallest shadow economy at 7.41 % of GDP; followed by Colorado, averaging 7.52 % of GDP, rounding out the three smallest shadow economies in the US West Virginia and Mississippi, on average, have the largest shadow economies in the US as a percent of GDP (9.32 and 9.54 %, respectively).
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Estimates provided by California’s Employment Development Department in analysis of SB 1185: http://www.leginfo.ca.gov/pub/11-12/bill/sen/sb_1151-1200/sb_1185_cfa_20120629_140323_asm_comm.html.
“California Officials vow to crack down on underground economy,” by Marc Lifsher, L.A. Times, December 09, 2011.
See, for example, the Fiscal Policy Institutes’ study, “The Underground Economy in the New York City Affordable Housing Construction Industry,” April 17, 2007. (http://www.fiscalpolicy.org/publications2007/FPI_AffordableHousingApril2007.pdf).
The shadow economy has many synonyms. Among them are: underground, informal, extralegal, black market, hidden, parallel, System D, cash economies.
This is a common claim in the literature, though one that is highly debatable. Consider, for example, the electricity consumption variable that will be used as an indicator in the analysis: electricity consumption statistics will measure electricity used for legal and illegal market activity, such as manufacturing marijuana.
District of Columbia is excluded from the study.
See, for example, Kucera and Roncolato (2008) for an extensive survey of studies that examine labor market regulations against shadow economy outcomes.
For example, a minimum wage requirement of $2 an hour in New York will have little impact, but for a developing nation, it might remove most potential workers from the effective workforce. The same idea holds, though in a narrower range, for jurisdictions within the United States (Ashby et al. 2011).
Though individuals should have the right to organize for the purposes of collective bargaining, laws and regulations governing the labor market often force workers to join unions when they would rather not, and make it difficult to escape unionization where coercion can most easily be employed (Ashby et al. 2011).
See the appendix for an abbreviated quote of the Census Bureau’s Classification Manual entry for “Protective Inspection and Regulation.” A definition of purpose, examples, and exclusions are provided.
In other cases (particularly cross-country studies) GDP per capita is used as a control determinant for the shadow economy size. It is useful in this sense when considering nations that are relatively heterogeneous—developing nations, and combined observations of high-income and developing nations, are examples of such a mix. Buehn and Schneider (2012), for instance, alternate their use of GDP per capita as a determinant and an indicator variable depending on the mix of economies they are considering in various specifications. For their specifications that consider developing, transition, and all countries in the study—developing, transition, and high income—they use GDP per capita as a determinant. When they consider only high-income OECD countries in their specifications—relatively homogeneous countries—they use GDP per capita as an indicator. Following Buehn and Schneider (2012), I consider real GDP per capita here as an indicator.
A condition typical of simultaneous equations models is the singularity of the regression matrix. A similar condition arises for the error covariance matrix as the result of the normalization requirement for the measurement equation (Chaudhari et al. 2006). For more on simultaneous equations modeling, see e.g., Bollen (1989).
Excluded from Table 2 are specifications which are regressed using only available data. Though regressions of this sort are run on a reduced number of observations, they yield no measurable difference in terms of economic and statistical significance of the coefficient estimates.
I exclude regression results for those that include sole proprietorship from Table 2 since goodness of fit statistics reveal that, when sole proprietorship is included, the model does not provide any useful information (χ2 statistic is 38.49; RMSEA is 0.056 with a p close value of 0.285).
As a robustness check, I also estimate shadow economy size using estimates from specifications 2 and 3. The results, whether generated by specifications 1, 2, or 3 are highly correlated (a basic correlation matrix reveals no less than a 0.82 correlation between alternatives). Moreover, the rank order of shadow economy size (reported in Table 3) does not change in any meaningful way; for example, Delaware, Oregon, and Colorado maintain the smallest shadow economies, while West Virginia and Mississippi have the largest, respectively, in any case. Additionally, results from a specification not reported here that include all causal variables except Government Consumption as a percent of GDP yield similar results (highly correlated at 0.99 to those used to estimate shadow economy size), though the model is a weaker fit (by RMSEA and Chi test stats) than those reported.
The state-level variables are standardized based on the national averages of the data.
Schneider (2010b) estimates are calibrated using results from a cash demand approach. One notable criticism of the MIMIC model is its use for constructing a cardinal time series of shadow economy size. Since the MIMIC model can only produce an ordinal time series index, cardinal calibration requires use of past shadow economy estimates, ultimately dependent on other methods such as electricity consumption and currency demand. For additional discussion of both advantages and disadvantages of the MIMIC model approach to estimating shadow economy size, see the Schneider (2010a) “Appendix”.
For a detailed explanation of the calibration procedure with examples, see the “Appendix” (A.2).
The authors also offer a link to the manual: http://www.census.gov/govs/www/classfunc66.html.
References
Alderslade, J., Talmage, J., & Freeman, Y. (2006). Measuring the informal economy: One neighborhood at a time, A discussion paper prepared for The Brooking Institution Metropolitan Policy Program.
Ashby, N., Bueno, A., & McMahon, F.; with Deborah Martinez. (2011). Economic Freedom of North America 2011, Fraser Institute.
Barrett, P. (2007). Structural equation modelling: Adjudging model fit. Personality and Individual Difference, 42, 815–824.
Bollen, K. (1989). Structural equations with latent variables. New York: Wiley.
Browne, M., & Cudeck, R. (1993). Alternative ways of assessing model fit. In K. Bollen & J. Long (Eds.), Testing structural equations (pp. 136–162). Newbury Park, CA: Sage.
Buehn, A. (2011). The shadow economy in german regions: An empirical assessment. German Economic Review, 13(3), 275–290.
Buehn, A., & Schneider, F. (2012). Shadow economies around the world: Novel insights, accepted knowledge, and new estimates. International Tax and Public Finance, 19, 139–171.
Campbell, N., Fayman, A., & Heriot, K. (2010). Including U.S. state government regulation in the economic freedom of North America Index. The Journal of Private Enterprise, 25(2), 165–186.
Chaudhari, K., Schneider, F., & Chattopadhyay, S. (2006). The size and development of the shadow economy: An empirical investigation from states of India. Journal of Development Economics, 80, 428–443.
Dell’Anno, R., & Solomon, O. H. (2008). Shadow economy and unemployment rate in USA: Is there a structural relationship? An empirical analysis. Applied Economics, 40, 2537–2555.
deSoto, H. (1989). The other path: Economic answers to terrorism. New York, NY: Basic Books.
deSoto, H. (2000). The mystery of capital: Why capitalism triumphs in the west and fails everywhere else. New York, NY: Basis Books.
Feld, L., & Schneider, F. (2010). Survey on the shadow economy and undeclared earnings in OECD countries. German Economic Review, 11(2), 109–149.
Frey, B. S., & Weck-Hanneman, H. (1984). The hidden economy as an ‘Unobserved’ variable. European Economic Review, 26, 33–53.
Johnson, S., Kaufmann, D., & Shleifer, A. (1997). The unofficial economy in transition. Brookings Papers on Economic Activity, 2, 159–239.
Kucera, D., & Roncolato, L. (2008). Informal employment: Two contested policy issues. International Labor Review, 147(4), 321–348.
Lacko, M. (1998). The hidden economies of visegrad countries in international comparison: A household electricity approach. In L. Halpern & C. Wyplosz (Eds.), Hungary: Towards a market economy. Cambridge: Cambridge University Press.
Lacko, M. (1999). Electricity intensity and the unrecorded economy in post-socialist countries. In E. L. Feige & K. Ott (Eds.), Underground economies in transition. UK: Ashgate Publishing Ltd.
Lacko, M. (2000a). Do power consumption data tell the story?: Electricity intensity and hidden economy in post-socialist countries. In E. Maskin & A. Simonovits (Eds.), Planning, shortage, and transformation: Essays in Honor of Janos Kornai. Cambridge, MA: MIT Press.
Lacko, M. (2000b). Hidden economy: An unknown quantity?: Comparative analysis of hidden economies in transition countries, 1989–1995. Economies in Transition, 8(1), 117–149.
MacCallum, R., Browne, M., & Sugawara, H. (1996). Power analysis and determination of sample size for covariance structure modeling. Psychological Methods, 1, 130–149.
Neuwirth, R. (2011). Stealth of nations: The global rise of the informal economy. New York, NY: Random House.
Schneider, F. (2010a). The influence of public institutions on the shadow economy. The Review of Law and Economics, 6(3), 441–468.
Schneider, F. (2010b). The influence of the economic crisis on the underground economy in Germany and the other OECD-countries in 2010: A (Further) Increase, Working Paper.
Schneider, F. (2012). The shadow economy and work in the shadow: What do we (not) know? IZA Discussion Paper No. 6423.
Schneider, F., & Buehn, A. (2009). Shadow economies and corruption all over the world: Revised estimates for 120 countries. Economics: The Open-access, Open assessment E-journal, 1.
Schneider, F., Buehn, A., & Montenegro, C. (2010). New estimates for the shadow economies all over the world. International Economic Journal, 24, 443–461.
Schneider, F., & Enste, D. (2000). Shadow economies: Sizes, causes, and consequences. Journal of Economic Literature, 38, 77–114.
Steiger, J., & Lind, J. (1980). Statistically-based tests for the number of common factors. Paper presented at the annual meeting of the Psychometric Society, Iowa City, IA.
Torgler, B., Schneider, F., & Schaltegger, C. (2010). Local autonomy, tax morale, and the shadow economy. Public Choice, 144, 293–321.
Acknowledgments
I gratefully acknowledge financial support from the Charles G. Koch Foundation, and the European Social Fund’s Doctoral Studies and Internationalisation Programme DoRa. Additionally, I would like to thank Andrew Young for his insightful comments and suggestions, many of which influenced my thinking during the construction of this paper.
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
1.1 Census Bureau’s Classification Manual entry for “Protective Inspection and Regulation,” [abbreviated by Campbell et al. (2010)]
“DEFINITION: Regulation and inspection of private establishments for the protection of the public or to prevent hazardous conditions NOT classified under another major function.”
“EXAMPLES: Inspection of plans, permits, construction, or installations related to buildings, housing, plumbing, electric power plant sites, nuclear facilities, weights and measures, etc.; regulation of financial institutions, taxicabs, public service corporations, insurance companies, private utilities (telephone, electric, etc.), and other corporations; licensing, examination, and regulation of professional occupations, including health-related ones like doctors, nurses, barber, beauticians, etc.; inspection and regulation or working conditions and occupational hazards; motor vehicle inspection and weighing unless handled by a police agency; regulation and enforcement of liquor laws and sale of alcoholic beverages unless handled by a police department.”
“EXCLUSIONS: Distinctive license revenue collection activities…; regulatory or inspection activities related to food establishments or to environmental health…; motor vehicle inspection, liquor law enforcement, and other regulatory type activities of police agencies…; regulatory and inspection activities related to other major functions, such as fire inspections, health permits, water permits, and the like…” (pp. 170–171)Footnote 23
1.2 Calibration procedure
The data used in MIMIC model estimation are first standardized relative to their national average. This means that MIMIC model estimates provide information about the latent variable (shadow economy size) relative to the national average shadow economy size. In order to estimate annual shadow economy size, two steps are required: (1) construct an ordinal time series index of state-level deviations from the national average shadow economy size; (2) calibrate the index in order to provide a meaningful cardinal time series.
The following equation is constructed from the MIMIC model estimation of specification 1 in Table 2:
Applying the underlying standardized data to this equation yields (1), the index. For example, standardized data for the states Delaware, Mississippi, Oregon, and West Virginia in 1997 are:
State | GOV | TAX | CHG | LABOR | INS | REG |
|---|---|---|---|---|---|---|
Delaware | 1.96 | −2.05 | 0.63 | 1.11 | −0.99 | 0.52 |
Mississippi | −1.24 | 2.07 | 0.26 | −1.46 | 0.45 | 0.26 |
Oregon | 0.36 | −2.11 | 0.24 | −1.46 | 1.36 | 1.05 |
West Virginia | −2.30 | 1.44 | 0.91 | −2.56 | 2.70 | −0.01 |
Multiplying each data point by the relevant coefficient in the SHADOW equation yields the following:
State | −0.397*GOV | 0.206*TAX | 0.068*CHG | −0.069*LABOR | −0.200*INS | −0.054*REG |
|---|---|---|---|---|---|---|
Delaware | −0.74 | −0.42 | 0.04 | −0.08 | 0.20 | −0.04 |
Mississippi | 0.47 | 0.42 | 0.02 | 0.10 | −0.09 | −0.04 |
Oregon | −0.14 | −0.43 | 0.02 | 0.10 | −0.27 | 0.04 |
West Virginia | 0.87 | 0.30 | 0.06 | 0.18 | −0.54 | −0.03 |
A horizontal summation of the above table yields each state’s 1997 index number. Step (2) requires an exogenous estimate of national average shadow economy size for calibration. I use Schneider’s (2010b) estimate for the size of the US shadow economy in 1997 (and all subsequent years). Summing each state’s index number and Schneider’s national estimate provides the state-level estimate for shadow economy size in 1997:
State | 1997 index | % GDP Schneider’s (2010b) national average shadow economy size for the US, 1997 | % GDP state shadow economy size, 1997 |
|---|---|---|---|
Delaware | −1.04 | 8.9 | 7.86 |
Mississippi | 0.88 | 8.9 | 9.79 |
Oregon | −0.68 | 8.9 | 8.22 |
West Virginia | 0.84 | 8.9 | 9.74 |
Rights and permissions
About this article
Cite this article
Wiseman, T. US shadow economies: a state-level study. Const Polit Econ 24, 310–335 (2013). https://doi.org/10.1007/s10602-013-9146-7
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10602-013-9146-7