E-Government and Corruption: Examining the applicability of using e-government to reduce corruption in IndiaAshwiniBlockedUnblockFollowFollowingMay 24(Spring 2017)E-Government is the use of information and communication technologies, commonly referred to as ICTs, within the public sector in order to improve the operations and delivery of the sector’s services.
Using these ICTs in government agencies as well as within the institutions of education and research are intended to result in more efficient processes as well as a quicker and more honest dissemination of information to the public.
Seeing as technology has had the effect in other mediums, such as the news media or communication strategies, to increase transparency, it would only be logical to apply the same technology into keeping the government honest.
With the continuous advancement of technology in the modern era, not only is there more technology to choose from when it comes to implementing e-government, but not using that technology to increase efficiency and transparency within the public sector is a waste.
In both developing and already developed countries, governments are seeking to leverage the new and expanding technology of today in order to ensure more accountability within the government and other public sector organizations as well as in improving quality and speed of services towards citizens.
In developing countries in particular, however, the adoption of e-government has a reportedly high failure rate, which is unsurprising as ensuring robust performance by large-scale information systems is difficult even for countries with more advanced technological skills.
Nevertheless, e-government has been seen in recent years as the solution to many of the inherent problems governments have in serving their constituents, and even more so in developing countries where typically the population of the countries restricts public agencies in improving their operations due to resource constraints.
When it comes to saving costs, improving quality, speeding up response times, and increasing access to services, e-government has the potential to improve the efficiency and effectiveness of administrations while at the same time increasing transparency and reducing corruption.
That last potential feature of e-government, reducing corruption, is key.
In developing countries, like India, corruption runs rampant, adversely affecting the economy and the overall efficiency of the government.
In a study conducted in 2009 by Transparency International, it was found that about 40 percent of Indians had dealt with first hand experience of either paying bribes or having to use “connections” to get elementary jobs done in the public sector.
Now currently the largest contributors to this corruption arte entitlement programs and social spending schemes that are enacted by the government, where much of the problem comes from the lack of knowledge on what the schemes actually entail a well as the additional fees typically asked for during the process of going through these schemes.
In particular, the manifestation of corruption in India takes the form of excessive regulations, complicated tax and licensing systems, bureaucracy, and the lack of transparency, all of which might be reduced through e-government.
This is the relationship that this paper seeks to explore — the relationship between administrative corruption and e-government in India.
In this case we looked at administrative corruption focusing on the firm-level which covered what a firm would need to do in order to attain basic services such as electricity and water.
I also looked at e-government as a statewide measure of what was offered to citizens in order to examine what the probability of firm-level government corruption was given citizen-level e-government offerings.
Literature ReviewWhile there are many studies that examine e-government and corruption as individual components of a country, very few examine their impacts on each other, especially in a developing country like India.
Implementing e-government should provide a country and its citizens with benefits such as increased efficiency in various governmental processes, transparency, and anticorruption, which will in turn lead to more citizen engagement and participation in the community as most prior studies on the reduction of corruption indicate.
Kumar and Best’s study looks at the impact and sustainability of e-government services in developing countries because typically e-government projects fail due to design-actually or design-reality gaps.
They longitudinally examine one case where e-government was extremely successful on the short term in a village in Tamil Nadu, India, but failed over the long term.
Focusing on how the presence of village Internet facilities offering government services affects the rate at which these services can be obtained, they use both quantitative and qualitative methods.
The study runs an OLS regression model for examining associations between availability of internet-based e-government services and the applications received for various services.
This resulted in a positive impact on government service provision by e-government, manifested through savings in time, cost, and effort required to obtain services.
It also resulted in reduced corruption for the first year after which the e-government services failed due to the lack of adequately trained personnel, public leadership, and unfortunate power dynamics (Best and Kumar 2006).
This starting point examining the context of e-government and corruption through the changing lens of technological development was conducted in the early 2000s, and since then the availability and access to technology has changed dramatically, which would hopefully impact the success of e-government ventures in developing countries.
Jindal and Sehrawat conducted an exploratory study on the condition of ICT infrastructure and its accessibility for using ICT based services in developing countries.
Similar to Kumar and Best’s study, this study found that the reasons most e-governance initiatives fail are due to the availability of infrastructure and the failure in bridging design-reality gaps.
Jindal and Sehrawat, however, believe this can be solved using ICT in e-governance.
This study was conducted in April 2016 which is recent enough that the technological advancement or the spread of technological advancement should be relatively similar to what it is now.
This means that the availability of technology which lends itself to the availability of the potential for e-government should be markedly similar (Jindal and Sehrawat 2016).
According to the study, however, a significant number of people surveyed did not have Internet facilities that would allow them to access ICT based services.
This indicated that the initial hurdle to jump over is the lack of technology infrastructure particularly in more rural settings.
In contrast, a majority of the population while using TVs for gathering political information also are using cell phones as the latest phenomenon of communication.
This indicates that progress in terms of technology that will be conducive to e-government is possible and with the more advanced technology, e-government can become more user-friendly encouraging more people to use e-government facilities instead of physically visiting government offices.
Ionescu also looks at the impact that e-government can have on reducing corruption and enhancing transparency through an analysis of the impact of ICT governed programs and initiatives to curb corruption.
Her results converged with prior research on the potential of ICT for enhancing social capital and anti corruption as well as the use of information technology to control corruption (Ionescu 2013).
Sharma and Mitra focus their study more on the impact of corruption, specifically in the way firms in India are impacted by corruption and whether e-governance could be of any use there.
In their study, they build on the existing literature about the factors involved in corruption to determine the impact of bribe payment on firm performance as well as to test the determinants of bribe payment, essentially who bribes and why.
Noting that illegal transactions can take place because of a denial of certain rights such as when smaller firms pay bribes to secure what they should already be due under normal circumstances, this study examines the relationship between growth and corruption.
There were two questions asked in the study; the first led to five hypotheses on the impact of corruption on firm productivity, and the second led to three hypotheses on what determines corruption in firms.
They looked at firm performance in particular, utilizing the World Bank data from a survey conducted on Indian manufacturing firms in 2005.
The first question was estimated using alternative frameworks and led to the conclusion that bribe payments work as a tax on profitability of firms and provide incentives for inefficiency.
The second question used Probit regressions and led to the conclusion that both tax compliance and the size of the firm have positive roles in bribe payment.
Sharma and Mitra also found that policy obstacles and bureaucratic complexity increase the probability of a bribe payment and that as a policy implication, these constraints need to be removed.
In their study on the effects of corruption on the manufacturing sector, Kato and Sato looked at the state level in India, using conviction rates of corruption-related cases to measure the extent of corruption and examined the impact of corruption on overall productivity.
They found that corruption reduces productivity, particularly in respect to smaller firms (Kato and Sato 2014).
However, the data sources they used in their study came with some shortcomings such as the fact that the Central Statistical Office which holds these records has an editing policy which impacts data consistency.
In their 2014 study, Chandiramani and Khemani examine the role that e-government has played in India in the past along with its shortcomings in order to identify its failures and work towards a more functioning e-government policy.
In India, the aim as of 2014 is to make e-governance mandatory in all government departments which would in turn reduce personal interaction of the public with government officials and thus help reduce corruption.
The current initiative in place includes the Information Technology Act of 2000 at the national level.
which in addition to regulating the elusive cyberspace, also introduced charters under which government departments has to make clear their goals, standards, and venues for grievances.
In addition the Ministry of Technology also plays a role nationally to facilitate e-governance.
On a state level, many states including Tamil Nadu have made progress in their quests to attain e-government.
Tamil Nadu has computerized major departments with the objective of restoring public confidence and creating an effective relationship between the citizens and the government.
In the past, e-government has often failed due to the lack of insight among policy makers, lack of transparency in governmental dealing, and a lack of mechanisms ensuring accountability among government officers (Chandiramani and Khemani 2014).
This article provides background knowledge on the existence of e-government in India as well as the extent and limitations of it currently.
The idea of reducing corruption through it is what I am trying to examine along with the impact of technology on that reduction.
The literature indicates that while numerous studies have been done on e-government and corruption separately, none had taken them all into account in a quantitative manner, particularly not recently with the advancement and availability of technology today, which is the central focus of my research.
The question in this paper is whether the probability of firm-level administrative corruption of various types increases or decreases given the e-government opportunities offered in the state in which the firm is based.
The additional questions of whether the availability of technology impacts corruption and whether the perception of corruption reflects the presence of e-government were also asked.
Given the benefits of technology in increasing transparency as well as the fact that with e-government, there is less of a likelihood of person to person interaction which is where bribery comes into play usually, I hypothesized that the presence of e-government decreases the likelihood of corruption.
I also hypothesized that the availability of technology would decrease the probability of corruption and that the higher the perception of corruption, the less e-government would be present.
Research MethodsThe data used in this research originated in two different places.
One is The World Bank Enterprise Survey’s Manufacturing Module of India dataset from 2014 which essentially has information on 1487 firms.
According to the methodology of the survey, the sampling provides firms from small, medium, and large enterprises and the questionnaire itself is conducted privately by contractors to the top managers or owners of each firm.
This dataset is quite expansive in asking questions about every area of the manufacturing business to firms in twenty-two states from all regions of the country of India.
From this mass of data, I narrowed down the variables we would be looking at in this particular study to the ones most relevant to corruption and technology, which I will discuss in the following paragraphs.
The other source of data was from the individual state websites of each of these twenty-two states.
These websites had information on what types of e-government services are offered in each of these states, which I appended to the original World Bank dataset based on the state information provided.
Unfortunately, given that India is a developing country along with the fact that there is no centralized source of information at the moment, the data collected about individual states’ e-government services is subject to some question (for instance, whether or not all services were included, whether or not the website had been updated, whether or not what one state considered to be a reportable service was not considered the same by another state, etc.
That is something that was understood going into the study and should be considered in evaluating the results.
From the World Bank firm data, there were three sets of key variables used.
A set for corruption measure, a set for perceptions measure, and a set for technology measure.
The corruption measures are based on the answers to questions on the questionnaire given to firms about whether or not they had to provide servicers with a “gift” or a bribe in order to get basic services.
The tricky part for all of the corruption measures was that in order to answer the question asking whether or not during their application for a certain service an informal gift of payment was expected or requested, the respondents had to first answer whether or not they had applied for that particular service in the last two years.
Given that short time frame, the dataset was narrowed considerably; however, there were still enough data points for each of these measures to ascertain an answer.
The first six measures listed in the table are whether an informal gift was requested or expected for 1) applying for electrical connection 2) applying for water connection 3) receiving a construction related permit 4) tax official inspections 5) applying for an import license 6) applying for an operating license.
The last measure (corr_index) is a sum of the first six indicating a total level of this administrative corruption within the firms: this measure has the value of 1, 2, 3, or 4 (none have five or six).
The technology measures assess how much general technology is used by these firms on a day-to-day basis.
These are dummy variables that indicate whether or not an establishment uses email to communicate with clients or suppliers, whether the establishment has its own website, and whether or not the establishment currently uses cell phones for their operations.
The perception measures assess how the managers and owners of these firms (the ones taking this survey) feel about the level of corruption as well as the obstacles they face.
There are seven measures I used for this.
The first is the perception of to what degree these people felt that the court system is fair, impartial, and uncorrupted.
The second through sixth look at to what degree tax administration, the court system, business licensing and permits, political instability, and general corruption are obstacles to the current operations of the firms.
The last measure takes into account what the biggest obstacle is.
In the case of the last measure, the variable is a dummy used to indicate whether corruption (which includes licensing struggles, instability, court corruption, and tax corruption) is perceived to be the biggest obstacle or not.
The e-government measures that were used in this research once again came from the websites of individual states included in the World Bank dataset.
These are e-government services that are offered for the citizens of the state, which we are using here under the assumption that the firm-level corruption is indicative of base-level administrative corruption and therefore the e-government services at the base-level offer insight.
The full list of these measures is listed to the left.
These were all dummy variables that indicated the presence of lack of presence of these services.
In this case they indicated whether or not a state’s government issued certificates including those for birth, death, caste declaration; whether there was a mechanism for issuing and resolving complaints; whether information about grain prices and other market information was made available; whether taxes could be filed online; whether bills could be paid online; whether one could register to vote online; whether information about government bills and laws was available online; whether education options such as registering for state-wide tests or applying for public university had an online forum; whether the state sent out newsletters about the latest services; whether one could apply for a trade license online; whether land records were available on the internet; whether transportation bills along with application for licenses was digitized; whether one could apply for government employment online or seek unemployment; whether e-procurement was an option; whether a mobile platform existed; whether there was an official directory of government officials available online; whether blood donor status was available online; whether services for those below the poverty line was online; whether one could find out online about the court cases; whether a cooperative audit information was available; whether pension services could be found online; whether hospital safety records for a mother and newborn infant were digitized; whether services intended to help those in rural areas were online; whether states had a resident data hub; whether there was global information systems to map the land, whether water tracking could be done online; whether immigration status and registration could be checked online; and whether passport application could be done online.
In addition to these e-government service measures, I also generated a new variable to sum up the other measures.
E-Government and CorruptionIn order to measure e-government and corruption, I ran multiple probit models to measure the probability of corruption occurring given the predictors of e-government.
There were six initial probit models run for each of the six corruption measures looking at the probability of corruption in electricity, water, construction, tax officials, import licensing, and operating licensing given the sum of e-government measures (so examining whether the number of e-government services provided by a state impacted the presence of corruption in each of these six administrative fields).
Pr(corruption) = Φ(β0 + β1egov) with “corruption” taking on each of the six measuresAfter these models were run and their marginal effects were computed, I ran probit models measuring again the probability of corruption in electricity, water, construction, tax officials, import licensing, and operating licensing given each of the individual e-government measures as predictors.
This was run to figure out which of the e-government measures best predicted the probability of corruption.
Pr(corruption) = Φ(β0 + β1issuecert + β2complaint +β3… )I also ran an ordered probit model on the probability of getting a 1, 2, 3, or 4 (there were no instances of 5 or 6) in the corruption index given the total e-government index.
Even though the numbers themselves are close to each other, there is a large difference between experiencing corruption (the requesting of bribes) in just one area of running a firm and experiencing that corruption in four separate areas.
That’s why the ordered probit model is used to measure the individual probability of receiving a 1, 2, 3, or 4 in the corruption index given the number of e-government services offered.
Technology and CorruptionFor the second part of my research analysis I looked at the probability of corruption overall given the use of technology by the firm in question.
The purpose of this portion of my research is to measure whether the use of email, cell phones, or websites by a company increases or decreases the probability that the same firm would experience corruption.
An ordered probit model was used to measure the probability of scoring a 1, 2, 3, or 4 on the corruption index given the presence of email usage, cell phone usage, and website usage.
Perceptions and E-GovernmentIn this area of research, the focus was on how people living in these states (in this case working or running these firms) perceived the level of corruption given the number of e-government services present in their state.
A probit models was run to predict the perception measures of the dummy response variables (perceived biggest obstacle faced by firms) given the e-government index (how many e-government services are present).
Five oprobit models were run.
Each of the other five perception measures include different levels to which people perceived tax officials, courts, licensing, unstable politics, and corruption to be problems.
So these models predicted the probability of each of those levels given the e-government index for the state.
ResultsIn my analysis I found that most of the corruption found in the dataset and in Indian firms during the year was with bribery on the part of tax officials; bribery with regards to obtaining an operating license was a close second.
I also found that Haryana, followed by Assam and Maharashtra, was the state that provided its citizens with the most e-government services, whereas Jharkhand and Karnataka were the states with the heaviest bribe-related corruption.
E-Government and CorruptionAfter examining whether the number of e-government services provided by a state impacted the presence of corruption in the electricity, water, construction, tax officials, import licensing, and operating licensing through a probit model, I found no significant correlation between the the number of e-government services and the presence of corruption in any of the administrative fields other than that of tax officials.
To reiterate, this variable indicates whether or not a “gift” or bribe was asked for when tax officials came to visit the firm.
An increase in e-government leads to a predicted decrease in the probability of a gift or bribe being requested by tax officials at a significance level of p<0.
Of course this being a probit model the coefficient listed in the table above is not easily interpreted at face value.
After computing the marginal effects for this model, it was found that a one unit change in the egov_sum decreases the probability of tax official corruption by .
This means that increasing the number of services provided via e-government may lead to a decreased probability of a tax official asking for a bribe.
This result aligned with my hypothesis particularly when we take into account that the corruption of tax officials from a firm perspective has the potential to be very similar to the corruption of tax officials from a citizen perspective.
An official who requires firms pay a gift during inspections probably asks the same of citizens.
This is particularly interesting to analyze when we remember that one of the e-government services offered by certain states is the ability to file taxes entirely online.
The second set of probit models that were run were measuring the probability of corruption in the same fields as before; however, this time with the predictors being each of the individual e-government measures.
The above tables indicate the results from the water model (the probability of corruption in attaining water) and the construction model (the probability of corruption in licensing construction).
The electricity model was excluded from this study because it yielded no significant results.
Marginal effects were computed for all of these and considered in the results.
Construction ModelWith the water model, issuing certificates, market information,bill payment, education information, and the availability of land records are the significant predictors.
Only education information availability, however, is consistent with my hypothesis that e-government decreases the probability of corruption.
This could be due to the exceedingly small number of observations for this measure.
The computed marginal effects indicate that the presence of education information online leads to decrease in the probability of corruption by .
As can be seen in the above models, with construction corruption, issuing certificates, providing market information, e-filing taxes, bill payment, enrolling in the electoral roll, providing education information, publishing e-gazettes, issuing trade licenses, e-procurement, and providing an official directory of government officers all yielded significant results (providing employment information did not).
For all of the mentioned above, excluding the publishing of the e-gazette, the official directory, and e-filing taxes, the results indicate that the presence of these e-government service decreases the probability of a bribery request when licensing corruption.
Especially with bill payment, this would seem to make sense given that if construction bills could be paid online, there would be less interaction with those licensing construction in the first place, where a potential bribe could happen.
Operating License ModelThe above two models look at bribery when tax officers come to visit and bribery in obtaining operating licenses.
With the tax model, significant results were yielded for the presence of a complaint redressal system, the availability of market information, the ability to pay bills online, the availability of education services, the availability of land records, transportation services, the presence of an official directory, e-information on judicial services, and cooperative audits.
For all of these services except the cooperative audit and the presence of an official directory, the probability of having corruption with tax officials decreases with the presence of these e-government services.
The official directory makes another appearance as a significant positive relationship between the presence of this directory and corruption, which is a rather interesting statistic to consider given that one would think that having a record of officials would lead to less corruption.
On the other hand, having such a record could also lend authority to those officials asking for bribes and gifts.
Interestingly enough there was not significant evidence to support that e-filing of taxes would reduce corruption of tax officials, which is something that I would want to examine further.
With obtaining operating licenses, the significant explanatory variables seem to be the availability of market information and the ability to enroll in electoral roll.
Only the availability of market information online has a negative coefficient, indicating that the presence of market information decreases the probability of corruption in receiving an operating license.
Importing License ModelAs for the importing license model, the significant explanatory variables are again the availability of market information, the ability to enroll in an electoral roll, and the availability of education information.
With the import license model, however, none of these statistics are negative indicating that the availability of these e-government services is actually associated with an increase in the probability of corruption in receiving an importing license.
This result seems to be fairly different from the others, which could actually be a result of the relatively small number of observations for this particular measure of corruption.
Since the question asked to firms is whether or not they had applied for an importing license in the past two years, it can be assumed that many of these firms do not deal with imports or have been in business long enough that they have had that license for more than two years.
Either way, due to the small number of observations, this result should be taken with a grain of salt.
The results from the ordered probit model used to measure the individual probability of receiving a 1, 2, 3, or 4 on the corruption index given the number of e-government services offered are shown below along with the computed marginal effects.
An increase in the sum of e-government measures offered by one unit leads to a decrease in the probability of the corruption index being 1, 2, 3, or 4 by .
This also supports my hypothesis and indicates that the probability of corruption decreases with the addition of e-government services.
Technology and CorruptionBelow are the results from the ordered probit model examining the probability of different levels within the corruption index given the usage of email, cell phones, and websites within the firm.
The results for the usage of email were not significant, whereas those for websites and cellphones were (albeit at a lower significance level).
Below are the computed marginal effects of that ordered probit model.
The marginal effects show that if a firm has its own websites, the probability of them experiencing corruption decreases by .
0209 (going from 0 to 1), by 0.
0162 (going from 1 to 2), and by 0.
00386 (going from 2 to 3).
With cell phone usage the probability of corruption actually ends up increasing.
Something to consider with cell phone usage, however, is whether or not the cell phones are smartphones, which would indicate some level of Internet activity (like a website indicates).
E-government and PerceptionsUnfortunately after running the tests on perceptions and e-government, no statistically significant results were able to be obtained and are therefore not included in this study.
That part of my research was inconclusive with the data that I had.
ConclusionCorruption in India is an age-old problem and one that greatly hinders India’s economic ability and decreases the interest and trust that constituents have in their government.
This paper supports the conclusion that e-government, the use of ICTs in the public sector, has the capability to reduce this corruption particularly at a base administrative level.
The results in this paper were varied, but in general supported the conclusion that the existence and proliferation of e-government, particularly with the availability of market information and the ability to file taxes and pay bills electronically, lead to a reduced probability of corruption at the firm level, specifically with regards to paying bribes or “gifts” in order to get basic services done.
As for the connection between the usage of technology and corruption, the data findings in this paper are minimal given that the only predictors being tested were the use of email, websites, and cell phones.
My hypothesis was only partially correct as the company having its own website did decrease its probability of experiencing corruption (and its probability of experiencing multiple forms of corruption), but the company using cell phones actually increased the probability of experiencing corruption.
These results are inconclusive about the effect of technology on corruption; however, examining whether or not there is a difference in the results between simple cell phones and smartphones would be interesting as the presence of Internet might be the dividing factor between the website predictor and that of cell phones.
In terms of the connection between perceptions of corruption and e-government, my results were inconclusive given that none of the resulting statistics were statistically significant.
Given a wider data source, this will be interesting to look at in the future.
In order for these results to be tested further, a new survey amongst the governments of various Indian states must be done to ascertain what e-government services they provide exactly as at the moment this paper assumes that those listed on their websites are fact, when in reality those may be outdated.
Surveying the individuals who use these e-government services or the constituency in general about whether they on an individual level have been asked for bribes or gifts for various services would also be a logical next step as currently we are relying on firm data and operating under the assumption that that is a strong indicator of citizen data.
Going forward, examining the spread of technology in these regions would also be an important factor to consider.
E-government is rapidly changing as the technology providing it changes, so it is valuable to note whether the people of India have access to all the changing technology or whether the effects of e-government are more concentrated on areas where this technology is most accessible.
In the rapidly changing digital world, e-government has the potential to help increase transparency and reduce corruption within the governments of developing countries like India, and more research should be done into the relationship between the two particularly with context given to the advancement of new technology.
“Use of information technology to address institutional failure.
” In Corruption and Anti-corruption.
Bertot, John C.
, Paul T.
Jaeger, and Justin M.
“Using ICTs to create a culture of transparency: E-government and social media as openness and anti-corruption tools for societies.
” Government Information Quarterly, 2010, 264–71.
Bhatia, Deepak, Subhash C.
Bhatnagar, and Jiro Tominaga.
“How Do Manual and E-Government Services Compare?.Experiences from India.
” In Information and Communications for Development.
“Transparency and Corruption: Does E-Government Help?” Commonwealth Human Rights Initiative, 2003.
Chandiramani, Komal, and Monika Khemani.
“E-GOVERNANCE: EXPLORING CITIZEN’S BEHAVIOR IN INDIA.
” International Journal of Research in Commerce and Management 5, no.
5 (May 2014).
De Waldemar, Felipe Starosta.
“New products and corruption: evidence from Indian firms.
” Dynamics of Institutions and Markets in Europe, 2011.
“THE IMPACT THAT E-GOVERNMENT CAN HAVE ON REDUCING CORRUPTION AND ENHANCING TRANSPARENCY.
” Economics, Management, and Financial Markets 8, no.
2 (2013): 210–15.
— — — “THE ROLE OF E-GOVERNMENT IN CURBING THE CORRUPTION IN PUBLIC ADMINISTRATION.
” Economics, Management, and Financial Markets 10, no.
1 (2015): 48–53.
Jindal, Neena, and Anil Sehrawat.
“Status of User Centric E-Governance Practices in North India.
” ResearchGate, April 2016.
Kumar, Rajendra, and Michael L.
“Impact and Sustainability of E-Government Services in Developing Countries: Lessons Learned from Tamil Nadu, India.
” The Information Society, 2006.
“Framework of E-governance at the Local Government Level.
” In Comparative E-Government.
Sharma, Chandan, and Arup Mitra.
“Corruption, Governance and Firm Performance: Evidence from the Indian Enterprises.
” National Institute of Financial Management.