Abstract
This chapter presents the results of ten logistic regressions that mirror the ten dependent variables used as proxies to estimate informality in Paraguay in 2017. The ten measures are organized into three groups, (1) social benefits, comprising firms that offer employees medical care, social security, and paid vacations; (2) organizational practices, encompassing registered firms, firms that provide client receipts, unionized firms, and employee contracts; and (3) firm characteristics, comprised of the self-employed and limited number of employees. Summary tables are provided for each group. Confirmatory estimates from the 2018 encuesta permanente de hogares are provided.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
This is calculated as the absolute value of 1 minus β or |1–1.813| or 0.813 or 81.3%, and assumes all other variables are held constant. This is the same pattern we utilize for the remainder of the chapter.
- 2.
The likelihood of having health insurance is 50.3%, 123.1%, 128.7%, 242.2%, 274.0%, 531.6%, and 445.0% higher, respectively, for enterprises with 6–10 workers, 11–20 workers, 21–30 workers, 31–50 workers, 51–100 workers, 101–500 workers, and 501 or more workers, than own account (owner-operator) enterprises.
- 3.
See the proportional chance criterion at a2 + (1−a)2, where a equals the proportion of one group to the whole, or (0.671)2 + (1–0.671)2 = 0.558. A good model predicts 1.25 times better than chance or 0.698. The present model predicts 0.822, hence the model predicts better than 1.25 times that of chance.
- 4.
The likelihood of enrollment is 108.6%, 107.1%, 384.5%, 229.2%, 146.9%, and 168.1%, respectively greater in agriculture, manufacturing, utilities, wholesale and retail trade, transportation, storage, and communications, and real estate and finance in reference to those who work in services.
- 5.
See the proportional chance criterion at a2 + (1−a)2, where a equals the proportion of one group to the whole, or (0.763)2 + (1–0.763)2 = 0.638. A good model predicts 1.25 times better than chance or 0.800. The present model predicts 0.892, hence the model predicts better than 1.25 times that of chance.
- 6.
As compared to the Metropolitan Region, the Dynamic Border Region, the Less Dynamic Border Region, and the Region of Economic Take-off are 37.0%, 33.0%, and 39.4%, respectively, more likely to have a paid vacation.
- 7.
The likelihood of workers having a paid vacation is 50.5%, 71.1%, 119.5%, 173.2%, 267.2%, and 118.0%, higher, respectively, for enterprises with 11–20 workers, 31–30 workers, 31–50 workers, 51–100 workers, 101–500 workers, and 501 or more workers, than own account enterprises.
- 8.
See the proportional chance criterion at a2 + (1−a)2, where a equals the proportion of one group to the whole, or (0.582)2 + (1–0.582)2 = 0.524. A good model predicts 1.25 times better than chance or 0.655. The present model predicts 0.807, hence the model predicts better than 1.25 times that of chance.
- 9.
The likelihood of working in an establishment with a RUC is 66.5% (2–5 employees), 486.9% (6–10 employees), 13.2 times (11–20 employees), 14.2 times (21–30 employees), 58.4 times (31–50 employees), 21.9 times (51–100 employees), 77.7 times (101–500 employees), and 78.5 times (501 or more employees) higher, respectively, than own account enterprises.
- 10.
See the proportional chance criterion at a2 + (1−a)2, where a equals the proportion of one group to the whole, or (0.509)2 + (1–0.509)2 = 0.500. A good model predicts 1.25 times better than chance or 0.625. The present model predicts 0.878, hence the model predicts better than 1.25 times that of chance.
- 11.
Firms with 2–5 workers, 6–10 workers, 11–20 workers, 21–30 workers, 31–50 workers, 51–100 workers, 101–500 workers, and 501 or more workers, are 58.5%, 3.9 times, 9.9 times, 12.7 times, 25.4 times, 16.7 times, 24.5 times, and 79.3 times, respectively, more likely to provide customers with facturas than own account enterprises.
- 12.
See the proportional chance criterion at a2 + (1−a)2, where a equals the proportion of one group to the whole, or (0.518)2 + (1–0.518)2 = 0.493. A good model predicts 1.25 times better than chance or 0.616. The present model predicts 0.878, hence the model predicts better than 1.25 times that of chance.
- 13.
Firms with 2–5 workers, 6–10 workers, 11–20 workers, 21–30 workers, 31–50 workers, 51–100 workers, 101–500 workers, and 501 or more workers, are 65.3%, 47.2%, 45.3%, 21.9%, 25.8%, 41.4%, 29.1%, and 14.9%, respectively, less likely to have union or association employees.
- 14.
See the proportional chance criterion at a2 + (1−a)2, where a equals the proportion of one group to the whole, or (0.931)2 + (1–0.931)2 = 0.935. A good model predicts 1.25 times better than chance or 1.117. The present model predicts 0.934, hence the model does not predict better than 1.25 times that of chance. But no model would because of the skewness of union membership.
- 15.
See the proportional chance criterion at a2 + (1−a)2, where a equals the proportion of one group to the whole, or (0.645)2 + (1–0.645)2 = 0.542. A good model predicts 1.25 times better than chance or 0.678. The present model predicts 0.806, hence the model predicts better than 1.25 times that of chance.
- 16.
See the proportional chance criterion at a2 + (1−a)2, where a equals the proportion of one group to the whole, or (0.640)2 + (1–0.640)2 = 0.539. A good model predicts 1.25 times better than chance or 0.674. The present model predicts 0.818, hence the model predicts better than 1.25 times that of chance.
- 17.
See the proportional chance criterion at a2 + (1−a)2, where a equals the proportion of one group to the whole, or (0.624)2 + (1–0.624)2 = 0.530. A good model predicts 1.25 times better than chance or 0.663. The present model predicts 0.847, hence the model predicts better than 1.25 times that of chance.
- 18.
See the proportional chance criterion at a2 + (1−a)2, where a equals the proportion of one group to the whole, or (0.724)2 + (1–0.724)2 = 0.750. A good model predicts 1.25 times better than chance or 0.663. The present model predicts 0.837, hence the model predicts better than 1.25 times that of chance.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2019 The Author(s)
About this chapter
Cite this chapter
Pisani, M.J., Ovando Rivarola, F.G. (2019). Logistic Regression Results of In/Formality in Paraguay. In: Understanding the Determinants of Economic Informality in Paraguay. Palgrave Pivot, Cham. https://doi.org/10.1007/978-3-030-24393-7_4
Download citation
DOI: https://doi.org/10.1007/978-3-030-24393-7_4
Published:
Publisher Name: Palgrave Pivot, Cham
Print ISBN: 978-3-030-24392-0
Online ISBN: 978-3-030-24393-7
eBook Packages: Economics and FinanceEconomics and Finance (R0)