## Monday, November 28, 2022

### The litigant perspective upon courts

Pavithra Manivannan, Susan Thomas and Bhargavi Zaveri-Shah

How do we identify a well performing judiciary from one that does not perform well? The literature on this question has focused on two types of metrics: inputs such as the judge to population ratio, judicial budgets and physical infrastructure and outputs such as the number of resolved cases, time taken per case and the costs involved. An emergent literature focuses on the litigant's experience of the judiciary. This approach involves criteria that the litigant uses to evaluate their experience of the judiciary, which have been found to be different from those used by judges, legal practitioners and planners (eg., Tyler, 1984; Rottman and Tyler, 2014; Hagan 2018).

In India, there is a growing awareness for the judiciary to be more citizen friendly (example; example; example), which calls for a better understanding of what a litigant's expectations are when engaging with the judiciary. In a new working paper, we propose a measurement framework that focuses on the litigant's perspective. In order to construct the framework, we draw upon the literature to hypothesise what a litigant takes into consideration when she decides to take a dispute for adjudication at a court. These considerations are then translated into the metrics to be used, when designing an evaluative framework to compare courts with similar functions. When this framework is applied to data from the legal system, it becomes an information system which can generate quantitative expectations of the time and costs involved in the process of litigation, which can potentially guide the litigant on whether to litigate.

In designing such a measurement framework, we recognise that there cannot be a single set of metrics that can be applied equally to all courts. This is because different courts perform functions that vary substantially in complexity, type and processes. For instance, the evidentiary burden required to be followed in a criminal matter is different than that of a civil matter, and the prosecution is led by the state. Additionally, the intended relief to a litigant in different types of matters also varies. For instance, in a civil matter, the relief is largely limited to compensation, specific performance and/or damages from the defendant. On the other hand, in constitutional matters, the relief sought may involve directions to the state or lower courts. While there may be some common metrics that could be useful to evaluate different types of courts, a single set of metrics may make the evaluation framework over expansive or deficient for some types of courts. Therefore, in this paper, we limit the scope of our discussion and the resulting framework to courts that adjudicate contractual disputes.

#### Features of the proposed framework

Given the focus on contractual dispute resolution, we identify a list of five metrics from the literature which can be usefully applied by a litigant to evaluate the performance of a court. The metrics are independence, efficiency, effectiveness, predictability and access. Based on the multiple interpretations of each metric available in the literature, we present arguments that justify why we narrow down on one interpretation over another from a litigant's perspective. We then identify proxies that can be used in the Indian context to measure the performance of the chosen courts on the selected metrics. These make up the proposed framework to measure the performance of courts that adjudicate contractual disputes.

The metrics, and the proxies that can be meaningfully evaluated to assess the metric, and the description of each proxy are summarised in the Table below. Finally, in the paper, we also lay out the source of the data and the process in which the information on each of these metrics can be collected.

Table: Metrics for evaluating court performance on contractual disputes

Sr. No. Metric Proxy Description
1. Independence Procedural fairness Adherence to procedure
Distributive fairness Fairness and impartiality in judgements
2. Efficiency Timeliness Duration of disposed and pending cases
3. Effectiveness EnforceabilityRatio of sum recovered to the total sum awarded in court orders
4. Predictability Certainty of case trajectoryClarity on stages of the case
Hearing date certaintyCertainty on number of hearings per case
Ratio of substantive to non-substantive hearings
5. Access Monetary costsCosts of approaching the court to the litigant
ConvenienceEase and user-friendliness for litigants

There are two caveats to the measurement framework that we propose. First, we assume that the litigant assigns equal weights to each of these metrics in making her decision on whether to take a contractual dispute in court. This means, that the litigant values (say) independence as much as predictability. This is a simplification and may not necessarily hold in reality, and for each litigant. Second, we do not identify an optimal or ideal level of performance of the court on these metrics. For example, we do not attempt to identify an ideal duration for the disposal of a case or the optimal number of hearings or the optimal 'level' of independence. The aim of the proposed framework is simply to provide a transparent base of metrics about court performance that can be put together using publicly accessed data sources, that we believe matters to the litigant.

The public domain nature of the data used in the proposed framework, supports regular updates of the metrics. This, in turn, will facilitate a comparison of the performance of court adjudicating contractual disputes over time. If these measures can be calculated in a consistent manner across different platforms, these can provide the litigant with a relative performance evaluation that can allow her to decide when, if and how to avail of the justice delivery system with greater clarity and certainty.

#### Conclusion

While judicial under performance is an over used expression in both the academic literature and broader policy discourse on Indian courts, the absence of an evaluative framework exacerbates the ambiguity associated with this expression. Our literature review in this paper shows that what is measured in the context of courts largely depends on who is undertaking the measurement. By considering specific metrics that a litigant may attach priority to in her experience with the judiciary, this paper provides a foundation for rolling out regular evaluation exercises of courts adjudicating commercial disputes, and ultimately make judicial performance a more tangible and usable concept in India.

#### References

Hagan MD (2018). “A Human-Centered Design Approach to Access to Justice: Generating New Prototypes and Hypotheses for Intervention to Make Courts User-Friendly.” Indiana Journal of Law and Social Equality, 6(2), 199–239.

Rottman DB, Tyler TR (2014). “Thinking about judges and judicial performance: Perspective of the Public and Court users.” Onati Socio-legal Series.

Tyler, Tom R. "The Role of Perceived Injustice in Defendant's Evaulations of their Courtroom Experience." Law & Society Review, vol. 18, no. 1, 1984, p. 51-74.

Pavithra Manivannan is a senior research associate at XKDR Forum, Mumbai. Susan Thomas is Senior research fellow at XKDR Forum, Mumbai and Research Professor of Business at Jindal Global Business School. Bhargavi Zaveri-Shah is a doctoral candidate at the National University of Singapore.

## Saturday, November 05, 2022

### Learning by doing and public procurement in India

by Aneesha Chitgupi, Abhishek Gorsi, and Susan Thomas.

### Introduction

State objectives can feature a mix of make' (i.e. build an organisation, recruit people) vs. buy' (i.e. contract to a private person for the same task). While private persons are, in general, more efficient than a comparable government organisation, the buy' pathway is not a panacea when there are bottlenecks in state capacity to contract. Understanding how to achieve capacity in government contracting is one of the critical components of the overall question of enhancing state capacity in India.

How has this tradeoff evolved over time? Generally, when contracting capabilities go up, the fraction of work that is contracted-out goes up. Chitgupi and Thomas (2022) find that the government procurement spending has been at a steady 17 percent of the total expenditure across multiple years, which suggests a lack of movement on contracting capabilities.

What about cross-sectional variation in contracting capability between spending units? Sharma and Thomas (2021) finds that procurement expenditure varies significantly across ministries in 2016-2017, with a few ministries accounting for the larger share of procurement. This would, of course, largely reflect differences in the nature of their work.

In this article, we obtain insights into contracting capabilities by examining procurement under-spend across ministries and across time. We recognise that state capacity in procurement includes multiple dimensions: achieving a targeted level of quality at the minimum cost and on time. Our question is a more basic one about whether the spending happens at all. It requires a certain level of contracting capability in order to actually achieve the budgeted procurement expenditure. We then consider the role of learning by doing in procurement.

An insight in the field of state capacity is that it is extremely sticky; changes to state capacity arise only slowly (Kelkar and Shah, 2022). State capacity emerges from the organisation design (mandate, governance, organogram, processes, the feedback loops of accountability mechanisms). Repetition of the same task, over and over, under the influence of accountability mechanisms, induces the slow process of building capability. By this reasoning, we hypothesise that state entities which procure on a sustained basis will build the organisation design and capacity to doing procurement efficiently, and vice versa. The lack of sustained procurement activity will then predict procurement underspend.

### Approach

In previous work, we developed the methodology to estimate the fraction of procurement in the accounts reported in the Detailed Demand for Grants (DDG) document published by a ministry. The DDG for a given year reports two sets of numbers by item head: the amount budgeted for the year, the amount actually spent two years previously. So, the DDG in 2016 for (say) the Ministry of Health and Family Welfare will report the budgeted amounts for 2016, as well as the amount spent in 2014. These are reported by item head, and allows us to calculate the actual amount spent but with a lag of two years.

For our current analysis, we calculate what was budgeted and what was actually spent on procurement, across ministries and across years. We use these to estimate the procurement spending gap for the ministry as:

100 x (actual procurement spend - budgeted procurement) / budgeted procurement

A negative spending gap has the ministry spending less than planned. If procurement capacity is the dominant factor driving the ability to procure, then we would say that a ministry with good procurement capacity will have a procurement spending gap close to zero. Further, if the learning-by-doing hypothesis holds, then those ministries that budget consistently for procurement will have built procurement capacity, and be able to show procurement spending gap close to zero.

Other than procurement, a ministry spends on operational expenses like salaries and pensions, and on grants-in-aid and schemes. These do not depend upon specialised capacity as with the contracting capacity required in the case of procurement spending. As a benchmark, we also calculate the total spending gap as the difference between the actual and budget total spending of the ministry, as a fraction of the total budgeted spending. We use the total spending gap of the ministry to benchmark their overall spending capacity.

We calculate the total spending gap of the ministry as:

100 x (actual total spent - total budgeted) / total budgeted

For our analysis, we do the following steps:

1. Estimate the amount of procurement expenditure for each ministry for each year for which we are able to obtain the DDG. We use this to identify which ministries have consistently higher fraction of their expenditure in procurement.
2. Estimate the procurement under-spend for each ministry across different years.
3. Examine whether there is a link between the ministries with a higher fraction of procurement expenditure and a relatively lower procurement under-spend.

### Data

We collect the DDG published by the ministries of the Union Government to estimate the procurement and total spending gaps. The estimation procedure described in Sharma and Thomas (2021) requires the object head wise expenditure to be disclosed in the DDG. This level of information has been consistently disclosed across all ministries (other than for the Railways ministry which has a more complicated accounting structure) from 2014-2015 onward. The latest available DDGs are from 2020-2021.

There remains challenges in accessing the DDGs reliably for all the Union government ministries. In our exercise, we are able to locate the DDGs consistently from 2014-15 to 2020-21 for 10 ministries: Civil aviation, Coal, Environment, Finance, Health and Family Welfare, Home Affairs, Housing and Urban Affairs, Rural Development, Road Transport and Highways and Law and Justice.

For seven of these ministries, the DDGs contain ministry level object head wise expenditures directly. For three ministries, we had to sum up the expenditures across department DDGs. These include: the Ministry of Finance which is the sum of Economic affairs, Financial services, Revenue, Expenditure, and DIPAM. The Ministry of Health & Family Welfare is the sum of Health & Family Welfare, and Health Research. The Ministry of Home Affairs (MHA) covers Home Affairs, Cabinet, Police, and 8 departments for the UTs. The expenditure of Home Affairs, Cabinet and Police make up 74 percent of the MHA expenditure. In our analysis, we report the procurement expenditure for the MHA as the sum of these three departments, and do not include the UT departments in the analysis.

All rupee values used in these calculations are adjusted for inflation using the CPI index. The spending gaps are calculated for each year, from 2014-15 to 2018-19.

### Is there a link between consistent procurement and lower procurement under-spend?

We organise our findings into three figures, where each is in the form of a heat map. In these graphs, each row is a ministry, where the values going from left to right are those for 2014-2015 to 2018-2019. The deeper / stronger the colours, the higher the values. In all three figures, the order of the ministries stay the same: in descending order of estimated procurement expenditure in 2014-2015. The ministry with the highest procurement spending is on top.

In Figure 1, we present the estimated procurement spending of the 10 ministries. The Ministry of Road Transport & Highways had the highest estimated procurement spending (in Rs. billion) in 2014-2015, and the ministry of coal had the least. There is large variation in the procurement spend across the 10 ministries, as described in Sharma and Thomas (2021). Ministries such as Coal and Rural Development spend less than Rs. 1 billion on procurement and have consistently done so over last 5 years. Ministry of Road Transport & Highways is the top procuring ministry followed by Home Affairs, of which Department of Police is largest procurer. There is more variation across the years for the remaining ministries. For example, the ministry of Health & Family Welfare shows an increased estimated procurement expenditure in 2017-2018 compared to other years in this period. The Ministry of Law & Justice has a steadily increasing procurement expenditure during this period.

 Figure 1: Estimated procurement expenditure by ministry, 2014-15 to 2018-19

Figure 2 presents the estimated procurement spending gap from 2014-15 to 2019-20. Green stands for positive values (which is over-spending or under-budgeting). Red stands for negative values (which indicates under-spending or over-budgeting). Except for a few years, most ministries tend to under-spend. The exception is the Ministry of Road Transport & Highways, which is also the top procuring ministry.

 Figure 2: Estimated procurement spending gap (%) by ministry, 2014-15 to 2018-19

Figure 2 also shows a sharper amount of under-spending towards the latter part for some of the ministries. These include Finance, Civil Aviation, Rural Development and Coal. This contradicts the proposition that procurement capacity is systematically increasing over time.

These ministries also have a lower fraction of procurement in their spending. These ministries are less likely to develop procurement capacity. Among ministries with similar procurement spending patterns, the Ministry of Environment appears to be building procurement capacity, with a procurement spending gap closer to zero.

Lastly, Figure 3 presents the spending gap for total expenditure for these ministries across time. There is much more capacity among the ministries in managing to spend their overall budgets. Most of the heat-map shows colours mapping to values closer to 0. In fact, there is more evidence of over-spending than under-spending compared to Figure 2. The ministries of Finance and Rural Development have more instances of over-spending in their overall spending, while these are ministries which have estimated procurement under-spend.

 Figure 3: Total spending gap (%) by ministry, 2014-15 to 2018-19

### Alternative explanations

Underspend can be driven by other compulsions also. In a public choice theory worldview, state organisations would give the highest priority to wage and pension expenditures, and sacrifice procurement expenditures when faced with formal or informal budget constraints. This makes procurement budgets the most vulnerable to sudden cuts.

However, this argument should apply equally across all ministries. In fact, the ministries which do higher amounts of procurement should be the prime target for mid-year budget cuts. This argument predicts that under-spend should take place roughly everywhere, and maybe to a greater extent in the high-procurement ministries. The evidence, however, shows that under-spend is more prevalent in low-procurement ministries.

### Discussion

The main finding of this article is that ministries that tend to procure consistently, tend to have smaller procurement spending gaps. This is consistent with the idea that there is learning by doing, where doing procurement on a sustained basis gradually creates organisational capability for procurement.

If the key claim of this article is true -- that there is learning by doing in procurement -- this has a few interesting implications. In a department where there is low experience with procurement, the early years where procurement work begins will work poorly. In such a department, procurement under-spend is likely, with consequential failures in public expenditure programs and budgeting. In a department with low experience with procurement, a sudden jump in the procurement budget is likely to be associated with failure to spend. If the political leadership decides to push up the procurement spend of a department by three times, it would make sense to (a) increase the budget by only 20% and (b) initiate a capacity building program for procurement capability within that department.

Procurement is an expertise. No government organisation can sporadically do this well. It is an expertise which can be built, albeit over many years. In any government organisation, people and processes can be organised to focus on this expertise, and to devote time and effort on the entire pipeline of government contracting. The process of developing this capability can be accelerated by bringing in people with this specialised expertise. Strengthening the entire life cycle is required to successfully spend budget amounts. But this is only the beginning of success in procurement where government can contract to deliver quality projects efficiently, on time and at low cost.

### References

Aneesha Chitgupi and Susan Thomas. The make vs. buy decision of the union government, The Leap Blog. September 10, 2022.

Vijay Kelkar and Ajay Shah. In service of the republic: The art and science of public policy. Second edition, 2022.

Shubho Roy and Anjali Sharma. What ails public procurement: an analysis of tender modifications in the pre-award process, The Leap Blog, November 26, 2020.

Anjali Sharma and Susan Thomas. The footprint of union government procurement in India, XKDR Working Paper 10, November 2021.

Aneesha Chitgupi is a Research Fellow at XKDR Forum, Abhishek Gorsi is a doctoral candidate at the IGIDR and Susan Thomas is a Senior Research Fellowe at XKDR Forum and a Research Professor of Business at Jindal Global University. We thank Josh Felman, Sudha Krishnan, Ajay Shah and Anjali Sharma for feedback and comments.

## Thursday, September 22, 2022

### Preparing for financially self-reliant and accountable regulators

by Rishika Rangarajan.

### Introduction

Indian regulators are tasked with important functions in key sectors such as standard setting, supervising, and monitoring entities, enforcing standards, etc. Crucially, some regulators, namely the Real Estate Regulatory Authority, need to work towards developing and promoting their respective sectors. Achieving these goals involves employing technical and scientific capacity, engaging with relevant stakeholders, collating, and analysing sectoral data, etc. A regulator's ability to conduct these activities, independently and efficiently, requires adequate financial resources and flexibility.

In June 2022, the Insolvency and Bankruptcy Board of India (IBBI) published a Discussion Paper proposing a methodology for the regulator to become financially self-reliant. Currently, IBBI largely meets its budgetary requirements from government grants, with only 20% of its financial resources coming from regulatory fees. The paper observed that IBBI’s mandate being a resource-intensive one, requires “financial independence which allows the Board to have the required flexibility and human resources”. In the past, regulators such as the Securities and Exchange Board of India (SEBI) and the Competition Commission of India (CCI) have also asserted their financial independence, claiming that fees are a means to achieve self-reliance. While it may seem intuitive to allow regulators to charge fees to secure independence, there may be important consequences of such a fee-based model on the accountability of regulators as well (Rangarajan, 2021).

In this article, I discuss the implications of a fee-based model and the potential concerns that may arise lacking such a framework - for example, an unchecked ability to raise fees by regulators may allow the misuse of funds. This article: first, discusses the key sources of funding and the importance of financial capacity; second, provides a summary of the incomes of SEBI, Pension Fund Regulatory and Development Authority (PFRDA) and Food Safety and Standards Authority of India (FSSAI) between 2015 and 2020; third, discusses important case laws on fee-based models; and finally, concludes the need for a formal and codified process for raising fees by Indian regulators, in line with accepted international practices.

Designing financially self-reliant, independent, and efficient regulators requires careful deliberation, which has not yet been done. It is critical that a fee-based model be designed to factor in principles of independence, accountability, and transparency.

### Key Sources of Funding

Each parent Act creates separate accounts for regulators which will hold the grants of the government, fees and subscription charges and any other income such as interest, penalties, or disgorged amounts. Currently, the key sources of income for regulators are (i) grants-in-aid, (ii) fees and charges, and (iii) other funds. The process of raising money for each source is briefly summarised below:

1. Grants-in-Aid: To receive funds from the Appropriate government, parent statutes of regulators require regulators to prepare a budget with estimated receipts and expenditures each Financial Year. This budget is forwarded to the respective central ministry each Financial Year which will grant money from their annual budget, after due appropriation. This money is then taken from the CFI and is approved during the annual Union Budget presentation.

For example, each Financial Year, IBBI submits their estimated revenue, capital and expenditure to the Ministry of Corporate Affairs. The Ministry, after considering the Actual Revenue and Expenditure for the previous year, determines the budget for grants-in-aid during the annual Union Budget discussions. The funds, after approval, are credited to the IBBI Fund established under Section 222 of the IBBI Act 2016. All regulators follow a similar procedure to receive grants from the Appropriate government.

2. Fees and charges: Regulators raise money through fees from regulated entities to conduct services such as registration, licensing, granting approvals, and other such activities. Commonly, regulators impose three types of fees: flat fees, fees based on the value of the transaction, and fees based on the nature of the transaction.

The Act does not prescribe any process to calculate the quantum of fees and regulators have the flexibility to determine the required fees. Regulators issue regulations that prescribe the quantum procedure for collecting fees from regulated entities. These are laid before the Parliament. Currently, there is also no requirement for regulators to publicly disclose the rationale for imposing prescribed fees on regulated entities.

3. Other funds: Regulators can also invest their funds and receive interest on such investments. Other sources of money include penalties, donations, income from publications, interest from deposits, income from the sale/disposal of assets, etc. In addition, some regulators have a separate fund that holds other incomes.

PFRDA can establish a separate Subscriber Education and Protection Fund which holds grants and donations received, interest on investments made and penalties imposed by the authority. Similarly, SEBI credits all amounts disgorged to an Investor Protection and Education Fund. Regulators can also own capital assets and hold separate capital/corpus funds and earmarked/ endowment funds which are reflected in their Balance Sheet.

### The importance of financial capacity

The mandate of regulators is resource intensive. Some of the key expenses regulators incur include: (i) establishment expenses for salaries, wages, allowances, and other such expenses for employees; (ii) administrative expenses consisting of expenses on rent, electricity and water, vehicles, stationary, etc; (iii) grants and subsidies given to institutions or individuals (for ex., FSSAI offers grants to strengthen Food Testing Labs in states); and (iv) expenses for developmental activities including monitoring and supervising their sector

OECD’s Best Practice Principles for Regulatory Policy highlights that funding is one of the important pillars of regulatory authorities. Financial capacity is not only linked to efficiency, but also independence. It ensures that regulators have sufficient funds to conduct their activities and remain independent from any external factors, including the government or private sector.

Despite the early optimism around regulators, they often face similar capacity issues as the State. In 2018, the FSSAI cited that financial constraints have led to failure in upgrading their food safety mechanism. FSSAI sought a ‘quantum jump in budgetary allocation’ pointing out that counterpart organisations in other countries have a much higher proportionate budget.

On the other hand, SEBI is an entirely self-reliant regulator raising over Rs 800 crores through investments and fees in the Financial Year 2021 with a surplus of close to Rs 200 crores. Although there have been calls for transferring SEBI’s surplus funds to the CFI, there is nothing in the mechanism preventing regulators from raising fees that go above their budgetary requirements. This goes against notions of accountability and transparency - principles that form the bedrock of any public institution. Although a fee-based model allows regulators greater flexibility by avoiding delays and complications arising from the CFI disbursement process, it lacks processes that ensure accountability that come with a grant or tax-based regime.

### Comparing incomes of regulatory authorities

To understand the role of fees and grants in the financial capacity of regulators, I have considered the income sources of three regulatory authorities SEBI, PFRDA and FSSAI, between 2015 and 2020. Income accounts of regulatory authorities provide us with a summary of the two key sources of funds: grants and fees/charges.

Evidently, the three regulators largely rely on either of the two sources (i.e., grants and fees) while the proportion differs for each regulator. For example, while SEBI does not rely on any grants from the central government, PFRDA and FSSAI, through the five-year period, largely rely on grants. Both FSSAI and PFRDA’s reliance on fees and subscriptions fluctuates through the period.

SEBI was also able to raise a sizable amount of money through income and interest earned which includes deposits in banks and other institutions; interest through loans provided to employees and interest from brokers.

SEBI makes available their board meeting minutes on their website, which includes the meetings where they decide the quantum of fees. These meetings reveal some of the contexts for why and how SEBI charges fees and the methodology they follow to decide the quantum. For example, in 2017, SEBI reduced the fees payable by brokers by 25% taking into account the projected income and expenditure for the subsequent three financial years and reducing the overall cost of transactions in the market. Similarly, in April 2020 they reduced the broker turnover fees and filing fees on offer documents to counter the challenges faced due to COVID-19.

PFRDA also makes available some of their board meetings - but, of the available minutes, none of them discussed any determination of fee matters. FSSAI does not make any board meetings publicly available and therefore there is no information on their fee determination.

### Moving towards fee-based models

Fees have emerged as an important source of income for regulatory authorities, being seen as their way to function independently of the government.

The Report on Financial Sector Legislative Reforms Committee (FSLRC Report) submitted to the Ministry of Finance in 2013 reviewed the legal and institutional framework of the Indian financial sector. While discussing recommendations to reform the regulatory ecosystem, the report raised the importance of maintaining regulators’ independence. Amongst the reasons put forth, the FSLRC Report stated that regulators funding itself through fees would create “operational efficiencies” and ensures that the stakeholders who are the beneficiaries of the relevant market will bear the cost of regulation rather than the public as a whole. Regulators can also achieve freedom from the government on pay, potentially facilitating the hiring of experts. Fees can empower regulators to maintain independence from regulators and enable them to take timely decisions.

Fees are also easier to raise as compared to taxes, the latter being an important source of revenue for the central and state governments. The process of raising taxes is codified in the Constitution of India under Article 265 - “No tax shall be levied or collected except by authority of law”.

The process to raise taxes follows a multi-step process which begins almost six months prior to the date of presentation. Each ministry is required to submit estimated receipts and expenditures to the Government of India for their financial year which is examined by the Ministry of Finance in consultation with the Union Cabinet or the Prime Minister. After a series of consultations and discussions, the budget is presented in the Lok Sabha, usually on February 1st every year (commonly known as the Union Budget most recently presented by the Finance Minister, Nirmala Sitharaman).

Fees, on the other hand, are used by government departments, local authorities, and regulators to raise money to cover the costs of any services rendered. Over the years, courts have differentiated fees from taxes, empowering regulators to use fees to fund their activities and services. Courts have recognised fees as a legal means to fund regulators' activities but highlighted the need for a fair correlation between the fee charged and the cost of services rendered.

### Differing standards of fees

In 2001, the Supreme Court of India considered the petition filed by stockbrokers challenging SEBI’s high registration fee charges. SEBI required stockbrokers to pay an annual registration fee based on their annual turnover over a period of five years. This is one of the earliest cases that dealt with regulators’ right to impose fees on regulated entities to fund their activities. The petitioners argued that the high fees were “excessive”, “unreasonable and arbitrary”. Second, they claimed that the fee is without the authority of law and is a tax guised as a regulatory charge. Finally, the levy has no nexus to the purpose for which the fee is collected and the demand for collection based on annual turnover extended over five years is arbitrary.

The Court rejected their arguments and found that SEBI does have the right to impose fees under the parent statute and therefore is authorised by law. The court did not consider the arguments on the quantum of the fees but held that regulators are not required to show a co-relatable quid pro quo. The court, however, refers to the Justice Mody Committee report which recommended preferable methods to calculate reasonable fees with SEBI in principle agreeing to implement them.

More recently, in 2020, insolvency professionals filed a writ petition, seeking the striking down of IBBI’s regulation charging ad valorem professional fees on them. They contended that there was excessive delegation, and the Act does not empower them to charge fees based on annual turnover or remuneration. They also raised that IBBI has not provided a quid pro quo to justify the charges. The Madras High Court ruled that regulators do not have to present a direct correlation between the fee earned and service rendered. In recent years, other regulators including PFRDA, CCI and PNGRB have also started to charge fees on an ad valorem basis.

As evidenced by the instances above, fees are specifically differentiated from taxes. Currently, taxes are the largest source of revenue for central and state governments with the process to determine and raise taxes set forth in the Constitution. Fees are becoming similarly significant to regulators. However, the same institutional safeguards are not put in place for regulators.

### Designing financially self-reliant regulators

To raise taxes, governments must go through a rigorous and intricate process which accounts for principles of independence, transparency, and accountability. This has been coded into the Constitution of India. A similar framework is lacking for regulators funded through fees and charges, raising some concerns. While flexibility is necessary for regulators, equally, checks and balances need to be formalised to prevent misuse of their powers.

An unfettered right to raise fees can have far-reaching consequences on the relevant sector. High fees can impact the market since they are often translated into costs to the public directly or indirectly. To avoid this, it is important to ensure that there is a reasonable nexus between the cost of the services rendered and the fees charged. The FSLRC report highlighted that regulators should “clearly explain the fees it is charging and demonstrate that the fee is not disproportionate to the cost for the regulator”. The OECD report on The Governance of Regulators stated that the funding processes of regulators should be transparent and efficient while protecting their independence and objectivity.

An international example of good practice in raising fees is the Financial Conduct Authority (FCA) in the United Kingdom which is funded entirely by the fees and levies from the firms they regulate. On their website, they explain how they calculate their annual fees and in addition publish an annual consultation paper which sets out its proposal on fees for the upcoming year and the model for calculating the various levies. The paper is open to comments from all FCA fee payers and businesses considering applying for FCA authorisation or registration. Similarly, the parliament of New Zealand also published a document on guiding principles for the levy of fees and charges.

Similar processes are lacking in the Indian context. Currently, regulators are not required to conduct consultations to determine fees nor required to disclose their justification of the fees to the public or the regulated entities. This makes it challenging for the government, regulated entities and the public at large, who indirectly bear the indirect burden of high regulatory fees to question and examine regulatory budgets.

### Conclusion: formalising transparency and accountability

It is argued that budgetary independence is related to the larger autonomy of regulators - regulators can determine their staffing, they can incur sudden or additional expenditures without immediate justification to the State, and it may also improve the quality of their operations by allowing investment in new technologies or requirements to upgrade internal processes.

Regulators that raise money need to be accountable to regulated entities and the public. When governments raise taxes, they must comply with certain constitutional and legal principles before deciding on the quantum. Principles that are enshrined in the Constitution. The same processes are not sufficiently imposed on regulators - their parent Acts do not provide any limitation to their right to raise funds through levies nor does it prescribe any requirements for transparency.

With an increasing number of regulators and increasing responsibilities imposed on them, their role in Indian governance is critical. In this context, the need for financial capacity cannot be denied but does the current process to raise funds by regulators ensure the necessary accountability? We need to consider creating a codified framework that sets out the above principles for self-reliant regulators.

### References

Kapur, D and Khosla, M (eds.). (2019). Regulation in India: Design, Capacity, Performance.

Burman, A., & Krishnan, K. (2019).Statutory regulatory authorities: Evolution and impact.

Burman, A., & Zaveri, B. (2018). Regulatory responsiveness in India: A normative and empirical framework for assessment. William & Mary Policy Review, 9 (2).

Sundaresan, S. (2018). Capacity building is imperative. Column titled Without Contempt in the editions of Business Standard dated August 2, 2018.

Report of the Comptroller and Auditor General of India. Union Government Accounts of the Union Government, No 44 of 2017.

Ministry of Finance, Department of Expenditure. General Financial Rules 2017.

Rishika Rangarajan is a Researcher at the National Law School of India University, Bengaluru

### Introduction

The Securities and Exchange of Board of India (SEBI) is one of the most powerful regulators in India. As the regulator of one of the world's largest stock markets by market capitalization, SEBI has a variety of enforcement tools at its disposal. These include the imposition of monetary penalties, license cancellation and pursuing criminal proceedings against violators. The law empowers SEBI to issue directions to intermediaries, and more broadly, to persons associated with the securities market. Such directions may be of a prohibitory nature, such as restricting companies from raising capital in the public markets, disqualifying persons from acting on the board of publicly traded issuers and restricting access to the capital market altogether. They may also be of a remedial nature such as disgorging illegal gains made by violators or directing restitution to wronged investors. The grounds for issuing such directions are wide.

How has SEBI used these enforcement powers over time? Has it prioritized enforcement against some kinds of misconduct over others? If yes, have the priorities stayed static or changed over time? Do certain types of violations consistently entail certain types of sanctions? How efficient are the enforcement proceedings in terms of the time taken, and what is the success rate for enforcing such sanctions? Unlike some Indian financial sector regulators, SEBI follows a due process before issuing such orders, involving the issuance of a show cause notice and a hearing and publishes each enforcement order passed by its officials systematically on its website. This transparency in enforcement allows us to establish some basic facts on securities laws enforcement in India over a long observation period. In a new paper, we analyse over 8,000 enforcement orders passed by SEBI over a span of ten years to answer some of the questions we mentioned above. In this article, we summarize the key findings of our work.

### Data description

In our study period beginning 1st January, 2011 and ending on 31st December, 2020, SEBI passed 9048 enforcement orders, of which we were able to sucessfully download and parse 8032 orders. We then analysed these orders, using text-mining software we designed ourselves, to arrive at some summary statistics on the frequency and type of enforcement undertaken by SEBI during the study period. To answer more detailed questions on the nature of enforcement, we manually analysed a stratified random sample of about 10% of these orders. The sample was drawn from the set of orders involving four regulations, which are most frequently enforced by SEBI (as per our data), namely, orders pertaining to fraudulent and unfair trade practices in the Indian securities market (FUTP), violations of the Insider Trading regulations, the Takeover Code and Broker regulations.

As mentioned above, the SEBI Act empowers SEBI to pass two types of orders, namely, orders imposing monetary penalties and orders issuing directions. Such orders can be issued against intermediaries, market participants, issuers of capital or persons generally associated with the securities market. Until 2019, monetary penalty orders could be passed only by adjudication officers and directions would be issued by whole time members of the SEBI board. With effect from 2019, the members of the SEBI board have also been empowered to pass orders imposing monetary penalties. In addition to these, the law also empowers SEBI to settle violations upon the payment of a settlement fee, without passing a guilty verdict against the violator. Basis this scheme of the SEBI Act, we categorize the enforcement orders in our data set into three categories shown in the Table. On an average, SEBI issues 250 enforcement orders with directions and double the number of orders imposing monetary penalties each year. The SEBI Act also empowers SEBI to initiate criminal prosecution against persons accused of having violated the SEBI Act or the regulations made by it, but we do not take account of this typology of enforcement proceedings in our study.

Table: Enforcement orders (2011-20)
Type of order Type of sanction Total^
Orders by Adjudicating officers Monetary penalties 4911 (61)
Orders by Chairperson/member Non-monetary sanctions 2484 (31)*
Settlement orders Settlement fee 637 (8)
Total 8032 (100)
^Numbers in brackets are a percentage of the total. *We estimate that not more than 30 orders may involve a monetary penalty.

As is evident from the Table, securities law enforcement is largely undertaken in India through monetary penalties, but the proportion of enforcement undertaken through non-monetary sanctions is not trivial. Settlements account for less than 10% of the total enforcement orders in our data. The annual distribution of these types of orders is shown in the Figure. The Figure shows that from 2018 onwards, there has been a sharp increase in the intensity of enforcement, with the number of monetary penalty orders nearly doubling from the previous years. The proportion of settlements has also increased over time, particularly after 2016. While the growth in the size of the market, an increase in the intensity of regulation and enforcement capacity are intuitive explanations for this jump, more precise, causal explanations require further research.

### Findings

SEBI draws its substantive powers from a set of three laws, over and above the SEBI Act, namely, the Companies Act, 2013 (and its preceding legislation), the Securities Contracts (Regulation) Act,1956 (SCRA) and the Depositories Act, 1996. While the Companies Act largely deals with the incorporation of Indian companies and the governance of their affairs, it also governs primary issuances, the requirements to be met by public offer documents and some aspects of the governance of listed companies. These matters under the Companies Act are administered by SEBI. The SCRA governs the conceptual definition of securities and securities contracts, regulates some types of securities contracts and governs the licensing and affairs of stock exchanges. The Depositories Act, 1996 deals with the regulation of depositories and depository participants. Under each of these laws, and in particular under the SEBI Act, SEBI has issued regulations defining the registration and reporting requirements for intermediaries, the kinds of misconduct that will elicit penalties, and so on.

We find that orders against fraudulent and unfair trade practices (FUTP) are the single largest group (15%), followed by orders dealing with violations of the provisions of the Companies Act (11%), insider trading regulations (10%) and the takeover code (9%). The enforcement actions (i.e. the number of orders) under the remaining regulations are few, with some of them having witnessed enforcement not more than once during the study period. Some of these seemingly rarely-enforced regulations, such as the regulations governing alternative investment advisors, are relatively new, which may explain why they do not appear more often in our data. However, others, such as the regulations governing venture capital funds, stock exchanges and clearing corporations, are older, but we see fewer orders issued under these regulations as compared to other regulations. Whether this is because the regulations themselves are not violated as frequently by market participants, or because SEBI chooses not to enforce them, requires further study.

To answer more specific questions of these enforcement orders, we manually analysed a random sample of 818 orders (approximately 10% of the total sample) from amongst the orders against the following types of violations: (1) FUTP, (2) insider trading, (3) violations of the takeover code and (4) violations of brokers' regulations. Some findings from this micro-study are summarised below:

1. Duration of the enforcement proceedings: The formal enforcement process at SEBI begins with the appointment of an investigating authority who investigates the facts and reports her findings to the SEBI board. If the findings are adverse, a show cause notice is issued to the accused by the adjudication officer (where the proposed sanction is a monetary penalty) or a whole time member of the SEBI board (where the proposed intervention is a direction). We find that the median time for the issuance of a show cause notice is a little more than three years from the date on which the violation was committed. Further, the median time from the date of issuance of a show cause notice to the date of an order imposing monetary penalties is a year and a half. It is a little more than two years for orders issuing directions. A regulation-wise analysis of the duration suggests no relationship between the complexity of the violation involved and the duration of the enforcement proceeding.
2. Subject and outcome of enforcement: A bulk of the enforcement actions are in respect of unregulated entities, that is, entities that are not SEBI-licensed intermediaries. This phenomenon could be attributed to the type of violations that are most often enforced against, namely FUTP and insider trading. Both these practices would likely involve traders and market participants that are not SEBI-licensed intermediaries. Further, in nearly 80% of the cases, SEBI found the person(s) guilty of all the violations that they were charged with, with a marginally higher conviction rate for unregulated entities compared to regulated entities. The conviction rate for violations of the Takeover Code is also marginally higher, compared to violations under the three sets of regulations. It is hard to comment on the optimality of this high conviction rate as these enforcement proceedings are undertaken and decided by SEBI officers themselves. All orders of SEBI, except those rejecting an application for settlement, are appealable to the Securities Appellate Tribunal (SAT). The rate of appeals and the outcome of appeals before the SAT could be a rough proxy to evaluate the optimality of this conviction rate and would be a good direction for further research.
3. Proportionality of sanction: We find a lot of variation in the amount of penalty levied across cases. While the median (i.e. typical) size of the penalty is in the range of Rs 5,00,000, the average is in the range of Rs. 57,00,000. This difference indicates that while there are few cases where large penalties are issued, the size of these penalties is very large compared to the typically-imposed penalties. One explanation that could account for this variation is the amount involved in the violation. The SEBI Act requires an Adjudicating Officer to take into account, among other factors, the amount of disproportionate gain or unfair advantage made as a result of the default or the amount of loss caused to investors as a result of such default. However, we find that in a vast majority of the cases in our sample (90%), the size of the violation was not calculated.

We similarly find a lot of variation in the orders that impose sanctions other than monetary penalties. Out of 118 such orders, 82 orders restricted the market access of the accused. The duration of such restrictions varied from 15 days to 4 years, and we could not discern any relationship between the duration of the restriction on the one hand and the violation or the purpose of the restriction on the other. Further, courts have repeatedly held that SEBI's direction making powers are remedial and preventive in nature, and not punitive. However, it is unclear at what point an order that operates to restrict market access starts to become punitive in nature, since none of the orders in our data clearly draw the line between remedial and punitive measures.

### Conclusion

In India, the field of securities laws is often studied from the perspective of a specific case, individual legislative amendments or specific judgements of courts. While such analysis is useful, a slightly different, more quantitative approach is necessary to gain a systematic understanding of the manner in which the regulator uses the wide variety of enforcement tools available to it, the manner in which it seeks to enforce against different kinds of misconduct and the efficiency of its enforcement functions. The consistent publication of easily accessible enforcement orders by SEBI on its website makes it possible to undertake such systematic research on securities laws enforcement in India. This paper is one such effort to begin developing more systematic knowledge on enforcement of private law in India.

The data used for this analysis can be found here. The data-set can be cited as Zaveri Shah, Bhargavi; Damle, Devendra (2022), "Securities law enforcement in India", Mendeley Data, V1, doi: 10.17632/ppdk9pzfdp.1.

Devendra Damle is an independent researcher. Bhargavi Zaveri Shah is a doctoral candidate at the National University of Singapore.

## Sunday, September 18, 2022

### The make vs. buy decision of the union government

by Aneesha Chitgupi and Susan Thomas.

### Introduction

Like every organisation, the government faces the question of make vs. buy in carrying out its functions. Should it recruit people, and take on management responsibilities of directly making goods/services? Should it enter into contracts with private persons who already specialise in this production? Each choice has a different cost when using public funds. Managers in government face a Markup in State Production, which is the inefficiency of the state in direct production, and the state has to deal with a Markup in State Contracting, which is its inefficiency when contracting for procurement (Kelkar and Shah, 2022).

This question needs to be set in some basic facts. How large are the magnitudes of make vs. buy in the Indian state? Over the years, is there a bias towards producing goods/services in-house? Is there a process of growing maturation where the buy ratio is going up, or have the spate of difficulties in government contracting of the recent decade led to a certain retreat from buy in favour of make? In this article, we collect evidence for the union government, to offer some answers to these questions.

### Methodology and data constraints

We use the methodology in Sharma and Thomas (2021) to estimate what the Union Government spends on procurement in a year. This uses the annual statement of accounts (Accounts at a glance) published by Controller General of Accounts (CGA). The books of accounts of government follows cash based system, recording only cash transactions (as opposed to accrual based system where transactions are recorded upon becoming payable).

The calculation of what was spent on procurement uses the expenditure under various `object heads''. This is only feasible from 2015-16 onwards, when the CGA started disclosing expenditure under object heads in their annual reports. In the work presented here, we undertake this measurement from 2015-16 upto 2019-20. We ensure comparability by converting nominal values to real using the CPI.

Our calculations differ from the methodology employed by Sharma and Thomas (2021) on three counts:

1. We only report the procurement expenditure of the Union Government, which includes only the expenditure under its ministries. We do not include expenditure in Central Public Sector Enterprises (CPSEs), which accounted for 3.5x the procurement of the entire Union Government.
2. We calculate procurement expenditure for the Railways Ministry using a combination of its annual reports and the capital expenditure reported in the CGA annual reports for the Ministry.
3. We include only those items that can be unambiguously classified as procurement for the Union Government.

### Results

Figure 1 shows the real actual spending of the Union Government for 2015-16 to 2019-20. The fraction of procurement spending has been quite stable at around 17-18 percent of the total expenditure. The remaining 82-83 percent of budget each year is spent on salaries and pensions and payments in the form of Grants-in-Aid, which are about 20-25 percent of the budget each year.

Figure 1: Union Government spending, with fraction of total procurement, inflation adjusted, 2015-2019

 Figure 2: Union Government procurement expenditure, with fractions of capital and revenue, inflation adjusted, 2015-2019

Source for both figures: Annual statement of accounts by CGA, Annual report of the Indian Railways, and authors' calculation.

While the share of procurement within total expenditure has been stable, there is a small shift in the composition of Union Government procurement. In Figure 2, we see that share of capital in total procurement increased from 53.7 percent to 57.5 percent between 2015-16 to 2019-20.

Table 1: Changes in share of total and procurement expenditure of Union Government as GDP (in %), 2015-2019

Total expenditure Procurement expenditure Capital procurement

/ GDP / GDP / GDP
2015-16 14.36 2.42 1.30
2016-17 14.37 2.50 1.38
2017-18 13.92 2.37 1.31
2018-19 13.55 2.40 1.39
2019-20 14.23 2.39 1.37

Source: Annual statement of accounts by CGA, annual reports of Indian Railways and authors' calculation.

How have these changed as a share of overall GDP? Table 1 shows that these fractions tend to be stable, whether it is the total spending by govt, or the spending on procurement or on capital procurement spending. There are some changes during 2017-18. This is likely a result of lack of maneuverability to reduce revenue expenditure, and an emphasis on items such as subsidies (RBI, 2020).

### Discussion

Government contracting is a critical part of state capacity that influences how the government chooses between make or buy in providing public goods. Given the more intractable problems faced by the Indian state in its attempts at recruiting people and producing internally, the possibility of contracting out to private firms is appealing. In the best examples, capabilities in public procurement have fostered innovation. But, this requires state capacity in government contracting. The evidence in India points to weak capacity in government contracting. Whether it is high levels of public litigation at courts (Mehta and Thomas, 2022; Damle et al, 2021), increasing levels of delayed payments at PSEs to vendors (Manivannan and Zaveri-Shah, 2019) and delays in project completion (Burman and Manivannan, 2022). Private parties are increasingly withdrawing from contracting with state (Mehta and Uday, 2021). This hampers purchase by the state as there is an inferior landscape of potential sellers.

A strategy of procurement reform could potentially put the Indian state on a path to higher capacity. The foundations of the field, of procurement reforms, lies in establishing basic facts:

1. How large are the magnitudes of make vs. buy in the Indian state? About 17% of the expenses of the union government work through buy, the remainder work through make.

2. Is the buy ratio something that is stable or does it fluctuate from year to year? The buy ratio is remarkably stable; it changes very little from year to year.

3. Is there a process of growing maturation where the buy ratio is going up? Or alternatively, have the spate of difficulties in government contracting of the recent decade led to a certain retreat from buy in favour of make? The buy ratio is highly stable; it shows no time trend. What has changed is a small shift towards a increasing spending for capital procurement.

This shows that, in real terms, the allocation between the make vs. buy choice of the union government has remained the same. But there has been a rise in the share of capital procurement spending. This tends to involve spending in projects with longer maturity, with greater risks to the project during the contract implementation and management phase being the underlying driver for procurement failure. Bottlenecks to resolving problems in this stage of procurement will become an important area to focus on as the government reforms public procurement rules and processes.

### References

Vijay Kelkar and Ajay Shah. In service of the republic: The art and science of public policy. Second edition, 2022, forthcoming.

Pavithra Manivannan and Bhargavi Zaveri. How large is the payment delays problem in Indian public procurement?. The Leap Blog. 22 March 2021.

Charmi Mehta and Susan Thomas. Identifying roadblocks in highway contracting: lessons from NHAI litigation. The Leap Blog. 13 July 2022.

Devendra Damle, Karan Gulati, Anjali Sharma and Bhargavi Zaveri. Litigation in public contract: some estimates from court data. The Leap Blog. 26 May 2021.

Pavithra Manivannan and Bhargavi Zaveri. How large is the payment delays problem in Indian public procurement?. The Leap Blog. 29 March 2022.

Perun. Defence economics, and the U.S. production advantage, YouTube, 31 July 2022.

Charmi Mehta and Diya Uday. How competitive is bidding in infrastructure public procurement? A study of road and water projects in five Indian states. The Leap Blog. 22 March 2021.

Anirudh Burman and Pavithra Manivannan. Timeliness in government contracting: Evidence from the country's largest metro-rail network. The Leap Blog. 12 August 2022.

Anjali Sharma and Susan Thomas. The footprint of union government procurement in India. XKDR Working Paper 10, November 2021.

Aneesha Chitgupi is a Research Fellow at XKDR Forum, and Susan Thomas is a Researcher at XKDR Forum. We thank Abhishek Gorsi for excellent research assistance, Josh Felman, Sudha Krishnan, Anjali Sharma and Ajay Shah for their inputs and comments.

## Thursday, September 08, 2022

### Position for researcher in the field of The Land Market

XKDR Forum is looking for a researcher to work on a project, on the land rights and land ownership by women. The project involves studying women's land rights, holding patterns and control over land resources through field and desk research.

XKDR Forum is a Mumbai-based inter-disciplinary group of researchers working in the fields of land, household and firm finance, financial markets, public finance management and public procurement. In these fields, the group engage in academic and policy oriented research, and advocacy. The new recruits will come into an active research program in the field. Our published work in the field of land include the following:

XKDR Forum is looking for one researcher with the profile described below.

#### Research Associate

As a research associate, you will work on project deliverables under the supervision of the Research Lead. The requirements for the role of research associate are: a background in law/economics and public policy, quantitative skills are desirable, two years of work experience. You must be comfortable in working in an inter disciplinary research environment with people from varying backgrounds such as economics, law, public policy and data science. You must be curious and passionate about research and must be willing to work on independent outputs as well as in teams.

The remuneration offered will be commensurate with your skill and experience and will be comparable with what is found in other research institutions.

Interested candidates must email their resume with the subject line: Application for "Research Associate" at XKDR Forum, to Ms. Jyoti Manke at careers@xkdr.org by 15th September, 2022.

## Friday, August 12, 2022

### Timeliness in government contracting: Evidence from the country's largest metro-rail network

by Anirudh Burman and Pavithra Manivannan.

### Introduction

Infrastructure projects in India are plagued by delays (MOSPI, 2022). Proposed explanations include failures of government contracting for public procurement (Singh, 2010, Sinha and Vatsa, 2021, etc.). In this article, we measure delays in the procurement process of the largest metro-rail network in the country -- the Delhi Metro Rail Corporation (DMRC) -- which is considered a successful project. It fares well on global ranks on some parameters such as network length and ridership. The early phases of this metro rail project have been lauded for timeliness in execution and contract payments (Expenditure Management Commission, 2016).

We look at two distinct datasets to obtain a birds eye view and a procurement-oriented view of the delays in DMRC. We find that DMRC is prompt in stages of the contracting processes for which we are able to find evidence, but that the overall project implementation suffers from time overruns. We put this knowledge together to obtain insights into government contracting.

### Our approach

As with contracts drawn between any two counterparties, government contracting is a pipeline that runs through four phases (Mehta and Thomas, 2022): (I) Contract specification and design, (II) Contract tendering and award, (III) Contract management and (IV) Contract closure. Flaws in government contracting shows up as inefficiencies in public procurement such as delays in infrastructure projects, which in turn, results in cost overruns, loss in revenues, vendor dissatisfaction and lack of competition when government wants to procure, and ultimately, deprives the public of the intended benefits.

We use three data-sets to understand the timeliness of government contracting in DMRC projects.

1. A data-set of the time taken by the various DMRC projects, sourced from the CapEx database published by the Centre for Monitoring Indian Economy (CMIE).
2. A data-set of tenders awarded by DMRC, hand-constructed from the 'Contracts Awarded' section of the DMRC website;
3. A data-set of payments made by DMRC, hand-constructed from the 'Vendor Payment Details' section of the DMRC website and from an RTI application.

The first data is sourced from the CMIE CapEx database. The CapEx database records the date of significant events for each project. We collect data for the three operational metro networks constructed by DMRC, that is, Phase 1, 2 and 3. This data-set consists of project level information such as, the date of announcement of the project, initial completion date, actual completion date and time overruns.

The second dataset is a hand-constructed data-set consists of tender level information for awarded contracts of DMRC, such as the department calling for tenders, nature of work, date of publication of Notice Inviting Tender (NIT), date of issue of letter of acceptance and value of the contract. This data-set covers this information for 892 tenders for a period of 5 years (2016-2020). DMRC categorises these tenders into 7 heads: Civil and Architecture Works, Electrical Works, Operations and Maintenance, Rolling Stock, Track Works, Signalling and Telecom and Property Development.

In addition, we categorise the contracts for IT services and housekeeping works as 'Miscellaneous' and the procurement done by DMRC for other metros in the country as 'For Other Metros'. The highest number of contracts were awarded for Operation and Maintenance works (623) and the least for Rolling Stock (2). Table 1 shows the typology of procurement undertaken by DMRC during our study period.

Table 1: Typology of DMRC Procurement
Category 2016 2017 2018 2019 2020 Total
Civil and Architecture Works 17 28 9 15 9 78
Electrical Works 5 1 3 5 13 27
Operations and Maintenance 0 65 167 200 191 623
Rolling Stock 0 0 0 2 0 2
Track Works 12 2 0 1 3 18
Signalling and Telecom 8 1 0 0 3 12
Property Development 0 5 3 10 5 23
Miscellaneous 4 14 23 8 13 62
For other metros 5 7 17 8 10 47

Our second hand-constructed data-set consists of monthly bill payment status of DMRC. Pursuant to government communication (vide D.O.18(18)/IFD/2019 dated 05.11.2019), DMRC uploads its monthly vendor payment details on its website since December 2019. This data gives us periodic information about the bill submission date and the bill payment date of DMRC vendors. Since the website does not archive its data we obtained our data partly from the DMRC website and partly vide an RTI application made for this purpose. Our data-set consists of 20,654 bills for the period between November 2019 to August 2021. The payment period is unknown for about 1,550 bills in our data-set, which we discard in our analysis.

We restrict our study to benchmark DMRC's performance against the timelines prescribed by its internal guidelines (DMRC Procurement Manual, 2016 and General Conditions of Contract, 2019) and the Central Government procurement guidelines (General Financial Rules, 2017 and the Manual for Procurement of Works, 2019). We do not employ a comparative analysis with other procuring entities for two reasons: One, availability of data in government portals such as the CPPP (Central Public Procurement Portal) and websites of the procuring entities are often sparse and sporadic. Second, a deeper understanding of the fundamental functioning, internal rules, processes and organisational structure of each entity is required for a meaningful comparison and it warrants a separate study.

### Findings: Time overrun in DMRC project implementation

In the CapEx data, we are able to see that, between 1995 to 2021, there were three projects announced, implemented and completed by DMRC. These are the Phase 1, Phase 2, and Phase 3 lines. These have been operational from 2006, 2011 and 2021 respectively. From this data, we are able to locate various timelines for all three phases, including the date on which the Phase was announced, to the date on which they were completed and commissioned for public use for fully operational metro lines. We calculate the time overruns as the difference between the date projected initially as the completion date for a Phase and the date on which it was actually completed and operationalised. These are presented as project delays in Table 2.

Table 2: Project delays (in months)
Phase 1 15
Phase 2 32
Phase 3 102
Source: CMIE Capex Database

### Findings: Timeliness in contract award by DMRC

High-income countries, countries with greater political accountability, and countries with greater economic freedom process public works procurement in a more timely manner (Djankov and Bosio, 2020). Each of these countries does infrastructure procurement following its own regulations to award contracts. In India, works procurement is guided by the Manual for Procurement of Works, 2019. According to Clause 5.6 in this manual, the time taken by Ministries and Departments from the date of opening the tender to the date of awarding of contract is 90 days.

We estimate the actual time taken by DMRC to award tenders (Table 3). This is calculated as the time taken from the date of opening of the tender to the date of issuing of the acceptance letter. We find that, on an average, DMRC takes 91-92 days to complete the tendering process.

Table 3: Time taken to award tenders (in days)
Year No. of tenders Average time taken
2016 51 101
2017 123 98
2018 222 103
2019 249 81
2020 247 89
Average 178 92

### Findings: Timeliness in making vendor payments by DMRC

Payment delays are endemic in public contracts in India. DMRC has sought to avoid payment delays by including provisions for both interim and final payments within its Procurement Manual and General Conditions of Contract, 2019 (GCC). Depending on the type of contract, payments may be made at different stages of the procurement cycle. At Clause 11, the GCC provides for set timelines for the scrutiny of invoices and payments to be made by the procuring entity:

• Interim payments: A contracting firm may apply to the respective project engineer of DMRC requesting for an 'interim payment certificate'. This certificate will be issued based on achieved milestones or prescribed payment schedule in the contract, if any.
1. Within 21 days of the request, the project engineer must issue the interim payment certificate specifying the amount due to the contractor.
2. DMRC is mandated to make 80% of the certified payment amount within 7 days of issue of the certificate.
3. The balance 20% is to be made within 28 days of issue of the certificate.
• Final payments: Once the project engineer certifies that the contractor has completed all his obligations related to a particular work, the contractor is entitled to apply for a 'final payment certificate' with the required supporting documents.
1. Within 28 days of receiving this request, the project engineer must issue the final payment certificate stating the final amount due.
2. DMRC is mandated to pay the amount certified in the final payment certificate within 56 days of issue of this certificate.

We look at the vendor payments data-set of DMRC to analyse the adherence to these timelines. We find that, on an average, DMRC takes about 4 days to clear its dues from the date of submission of the bill by the vendors (Table 4). For the data-set in our study, the maximum days taken by DMRC to make its payment is about a year.

Table 4: Time period for clearance of dues (in days)
Year No. of bills Average time Median time Minimum time
2019 1326 5 2 0
2020 9887 4 3 0
2021 7891 4 4 0
Total 19104 4 3 0

Payment delays by public sector enterprises in India to their vendors far exceeds their procurement values (Manivannan and Zaveri, 2021). We find that DMRC is an outlier in terms of maintaining payment discipline to its vendors, and in adhering to the timelines provided in its GCC.

### Discussion

The public discourse on government infrastructure procurement focuses on delays and time overruns being an indicator of poor government contracting. In this article, we have analysed the capability of a procurement-intensive public sector enterprise to keep up with its timelines in two stages, that is, in contract award and payments. We find that DMRC takes about 3 months to award a contract, and about 4 days to clear its payment dues to vendors. Regardless of this exemplary performance on awarding contracts and paying vendor dues, we also find that the overall project implementation by DMRC failed to meet scheduled timelines to complete. All three phases took a longer time than originally expected. In fact, we observe that the overall project time delays increased from the Phase 1 project to the Phase 3 project.

Executing infrastructure projects on time has been a continuous concern and challenge in India. Our analysis about DMRC timeliness in awarding contracts and in making payments provides evidence against the popular perception that public projects are delayed due to delays in decision-making by the public authorities and their inability to make timely payments. Instead, we speculate that these are because of other factors for overall project delays, some of which could be misaligned allocation of scope and risk in procurement contracts and poor contract management. A deeper analysis into each project, procurement practises, financial and institutional structure of DMRC may help in understanding the reasons for its timely performance in certain procurement processes and the potential causes of time overruns. These learnings can then be adopted by other procuring entities to achieve better procurement and project outcomes.

### References

Ministry of Statistics and Programme Implementation Infrastructure and Project Monitoring Division, 434th Flash Report on Central Sector Projects, January 2022.

Ram Singh, Delays and Cost Overruns in Infrastructure Projects: Extent, Causes and Remedies, Economic and Political Weekly, Vol 45, No. 21, May 2010.

PC Sinha and Ananys Vatsa, Delays in Project Completion in India, Indian Journal of Projects, Infrastructure and Energy Law, January 2021.

Erica Bosio and Simeon Djankov, Timely procurement of public works, World Bank Blogs, February 2020.

Department of Expenditure, Ministry of Finance, General Instructions on Procurement and Project Management, October 2021.

Expenditure Management Commission, Recommendations of the Expenditure Management Commission, December 2015.

Department of Expenditure, Ministry of Finance, Manual for Procurement of Works, 2019.

Delhi Metro Rail Corporation Ltd., General Conditions of Contract, November 2019.

Pavithra Manivannan and Bhargavi Zaveri, How large is the payment delays problem in Indian public procurement?, The Leap Blog, March 2021.

Charmi Mehta and Susan Thomas, Identifying roadblocks in highway contracting: lessons from NHAI litigation , The Leap Blog, July 2022.

Charmi Mehta and Diya Uday, How competitive is bidding in infrastructure public procurement? A study of road and water projects in five Indian states , The Leap Blog, March 2022.

Anirudh Burman is an Associate Research Director and Fellow at Carnegie India. Pavithra Manivannan is a Senior Research Associate at XKDR Forum and Chennai Mathematical Institute. We thank Susan Thomas for valuable comments and discussions.