Wednesday, December 9, 2015

e-Governance & GSTN SPV

A friend of mine from had asked me about GSTN SPV and its purpose. 

It was recently that cabinet approved GSTN to implement GST across India. Hence this can be a useful piece of info. 

Some thoughts:

1. You may want to refer to my previous post on SPV. 

2. In exams there may NOT be direct question on GSTN IT architecture as it will be too much details, but you never know ! However reading through it will give you understanding of how things are structured in govt.'s IT decisions or what is the flow in govt. IT implementation sphere. This will help you include it in some related questions. 

Also, this post will help aspirants with Management as one of their optional( under MIS subtopic) and also to some extent in GS .. especially around e-gov and my writeup around "need of hour" for e-Gov. 

Enjoy reading. 

P.S: Apologies for any grammatical errors. Post was written superfast !! 



National e-Governance Plan (NeGP) 

The NeGP vision is to “Make all Government Services accessible to the common man in his locality, through common service delivery outlets and ensure efficiencytransparency and reliability of such services at affordable costs to realize the basic needs of the common man”. 

1. Launched in May, 2006 to improve public service delivery 

2. Web enabled delivery of services & service levels - so everything over internet and through browsers.

3. Centralized Initiative, but Decentralized Implementation - emphasis on PPP

4. NeGP Includes two major parts to it: 

(i) 27 Mission Mode Projects (MMPs) (increased to around 31 now)

> GSTN is one such MMP

(ii) Core infrastructure components - Common Service Centres (CSCs), Data Centres and Wide Area Networks(also called SWAN)

Core Infrastructure Support of NeGP: 

1. Common Service Centres (CSC) 

> More than100,000 tele-centers in 600,000 villages.
> Broad band internet enabled connectivity
> Implementation through PPP

2. State Wide Area Network (SWAN) 

> Secured network for Government work
> Connecting State HQs ,District HQs, Blocks HQs
> Minimum 2 Mbps Broadband Connectivity 
[Note: If you check the National telecom policy released recently, this provision also features there. So it kinda aligns with the National Telecom policy as well. There is a lot of emphasis on increasing broadband density of this country. National Knowledge Network is one such initiative.]

3. State Data Centres (SDC) 

> State of art Data Centers at each of 35 States/UTs
> Housing all applications and databases
> e-Delivery of G2G, G2C and G2B services
> State Portals, State Service Delivery Gateways 

4. Capacity Building Scheme (CBS) 
[Note: This also forms a major thrust area of NIU, which we will talk below]
> Recruiting 400 Professional from market
> Constituting State e – Mission Teams ( SeMTs)
> E- Governance roadmaps and Programme Management

Problems with e-Gov in India

Till few years back, there was a need to articulate the requirement of e-gov., to emphasize or chalk out, what projects will be part of e-gov., bring the focus on service delivery and service levels, put in place supporting infra (SWAN, CSC, SDCs), to bring in monitoring mechanism etc. 

So, a lot of backend work had happened then, to bring about awareness about e-Gov, work with individual ministries and depts. to emphasize the need for service deliveries, change the whole paradigm of governance to one focused on delivering services and also to create the support mechanism in backend etc. As a result what has happened that all of these projects which were initially identified have now got money and lot of them have moved and got themselves sanctions and that is no small deal as it takes a lot of effort to get money sanctioned from govt. 
But why is it that today with so many projects having money and rearing to go across the country, why is it that we do not have all services delivered today through these infrastructure which is setup? Why the CSCs setup across country is not functioning? So there must be some gap in service delivery which we need to identify.

The gap is in execution

The point is that, it is one thing to go through concept documents, conceptualize projects, make presentations, held seminars and concerts etc and making people aware WHAT is to be done, but going forward, making that HOW happen has been a big GAP. And that is where the need to put in new structures and mechanism within the govt, at the backend, should be the focus area. 

Problem No.1 :
Today the only way in which govt. knows to get into IT services is to get into long IT infrastructure procurement processes. Every project takes around 1-2 yrs of procurement and the whole effort is spent on deciding whether to buy a SUN or a HP or an IBM computer(lol ), something which actually in itself will not deliver any service. You get the money but then you don’t know how to spend the money and when you look at the scale in which it has to be done(across country), it looks almost like an impossible task.
So we do not have a way to identify a process, scale it up and then roll it across the country. 

Problem No. 2:
Second is the problem in resources – people resources. Where are the people to do all e-gov initiatives we plan to undertake?? We talk about BPR, process rejig, capacity building, but where are the people? Govt. doesn’t have the mechanism to source them and to put them into project. 
So enabling system is not built inside the government to do large projects like e-gov. They are there in governments to do many many other things but not large projects like e-gov. 

Inside every dept that is executing a project, you need skills to conceptualize, roll out, execute, project maintenance, implementation. So you need capacity and by the time you are end of version 1.0, you need to rejig and introduce verison 2.0. So for this kind of system how do we get the human capacity?
[Note: Remember windows operating system versions - Windows 95, 98, Millenium, Xp, Visa, 7,8 etc and the Mac OSx versions of cheetah, puma, panthar, tiger, lion, leopard etc. So due to time and budget constraints, everything can't be built in one version. This is also not cost effective for any company. So we roll it out in versions. Also it is important from market share perspective that if you are late in rolling out ur newer versions, some of your competitor will beat you. Hence you constantly need to upgrade yourself. ] 

There has been talks about training people. But how can we expect govt. officers(most of whom dont have an iota of idea about IT) to undergo few weeks of training and then be able to start executing e-Gov project? So we need experienced people who have experience in delivering such systems. We need to attract talent. So how do we get skills inside the govt. US/UK government can tap skills from India and make then do work for their country but our own govt. cant attract those talent? Its because we lack the institutional mechanism by which we can attract those talents.Outsourcing to a private vendor model is understood, but how do we bring them inside the government.. make then write off files, make decisions?

So the government tomorrow should be able to go to a single window and say can you give me infrastructure to do this project, resources to do conceptualization, roll out, maintainence (all on demands) and then that mission leader or the e-gov champion will actually start doing the work. On the contrary, today if someone has to start doing an e-Gov delivery, it take minimum of 2-3 yrs before he gets to a point where he can start the delivery as until 2-3 yrs he is battling the system to get to a point where he can start the work.

Create an institution, preferably as 51/49(private/govt) [You will understand the significance of this ratio(51/49) when you see how GSTN is formed] which has the flexibility and private control to which people can go and look forward to getting some resources for execution and go ahead and execute. So unless you have a structure and get people and mechanism to do it, projects are really not going to happen. 

New institutions have to be prepared.

[Note: The concept of NIU is created exactly to meet these needs. We will talk about it when we discuss the GSTN SPV]

Problem No. 3:
Third thing is usage of technology like cloud etc which government can use to shorten the delay in execution of project. collective sharing of resources between centre and state so that there is no duplication of infrastructure and we do not spent money and resources in such non-value adding activities. 

So the first phase of WHAT has been done and the demand has been created for e-gov. The people are now waiting and watching HOW we will do that and unless these three problems are put in order, we cannot create deliver e-gov services. 

National Information Utility (NIU)

National Information Utility (GSTN SPV is a type of NIU, which we will talk in some time) is one such initiative addressing some of the problem listed above. 


Govt. offlate has realised that in its generic functioning, there is involvement of complex IT system development and there are some critical projects it undertook, which have immense transformative power and can change India’s growth trajectory. These five high priority projects are:
1. Goods and Services Tax (GST)
2. Tax Information Network (TIN)
3. Expenditure Information Network (EIN)
4. National Treasury Management Agency (NTMA)
5. New Pension System (NPS)

So there was a need to roll out complex IT systems, to achieve project objectives and sustain high levels of reliable performance. Hence, in its budget speech 2010-11, Finance Minister had announced setting up of "Technology Advisory Group on Unique Projects (TAGUP)". 

TAGUP had suggested setting up of NIU.

Why NIU?

Typically, IT projects supporting Govt functions are developed within Govt for small projects, or outsourced to a Managed Services Provider (MSP) or one or more vendors for larger projects. In the MSP/vendor model, the Govt Dept puts a project team in place, which carefully studies the business requirements and writes a Request for Proposal (RFP). An MSP/vendor is then selected through a competitive bidding process. 

Managing such projects has been a great challenge. Often, the implementation team has had to face serious problems due to lack of financial independence, inability to get the right personnel and retain them, technological obsolescence, lack of speed and productivity in implementation, lack of ownership on the part of the user community within the department, leading to cost and time overruns and failure to fulfill the requirements. 

[Note: Check the three major problems I listed above under issues with e-gov. ]

MY EXPERIENCE - Govt. or its entity doing IT outsourcing !!!
Usually what happens with any company(including govt.) who are looking to outsource their IT development work is that they come up with RFP (or sometimes RFQ - request for quotation or even RFI - request for info). This RFQ is then floated to the sales dept. of different IT services companies. So companies like TCS, Wipro, Infosys, Cognizant, IBM, Accenture etc comes into picture. This RFQ document lists all such questions which the company wants the IT vendors to answers. So questions can be as varied as "Q1. What kind of IT solution would you like to propose for implementing GST in India and why?" OR "q2. What is the corporate governance structure of your organization" OR "Q3. where all in India or World do you have offices and how many resources in each branch" etc .. 

The sales dept. (also called business dev. dept. ) of IT company then co-ordinates with its various other depts. and 
prepares an answer. Proposes the IT solution and also quotes cost for it. 

They then submit the RFP/RFQ/RFI before the due date. The outsourcing company then decides on the costs and other risks and chooses either one or multiple vendors. Now a days its a prevalence to choose multiple vendors and not give all tasks to one vendor. this reduces the dependency on one vendor and also hedges risks, should that one vendor go bankrupt. So say TCS can take development work and Cognizant can take testing while IT infrastructure can go to IBM etc. 

Once the vendors are decided, agreements are signed, also called MSA - Master Service Agreements and the work begins. The vendor usually provides the outsourcing company with a set of SLA (service level agreements) with which it promises to adhere to while delivering service. So things like .. if there is an issue with server, the usual resolution time will be 12 hrs etc .. 

Many times these IT vendors(since they are not part of govt.), faces huge challenges to get their proposal implemented. Govt. depts overseeing the IT work usually takes a lot of time to provide signoffs to the proposed IT architechture or other milestones in IT development (See Waterfall or RAD model of IT development, as it has various stages of development). 


So, a class of institutions called National Information Utilities (NIU) is setup to handle all aspects of IT systems for such complex projects. These institutions would work in the spirit of partnership with Govt, helping to overcome some of the challenges of a pure MSP/vendor model. They would participate in high-level design, specification of requirements, proof-of-concept studies, while strategic control is retained within Govt. For actual implementation, the NIU then contracts with vendors from the market for specialised services while being completely responsible to the Govt. for committed deliverables and service levels.

The NIU model would overcome the problems faced by the Govt in implementing the projects with in-house skills or through the MSP/vendor model.

Usually NIU model will work good for areas where you have:

> multiple levels of govt. like centre, state, local etc
> multiple govt. depts.
> multiple stakeholders
> Projects that require significant business process re-engineering to leverage IT

Hence GSTN was created as an NIU for implementation of GST across India.

The Goods and Services Tax (GST), which will replace the State VAT, Central Excise, Service Tax and a few other indirect taxes will be a broad-based, single, comprehensive tax levied on goods and services.

GST is based on a destination-based taxation system, where tax is levied on final consumption. It is expected to broaden the tax base, foster a common market across the country, reduce compliance costs, and promote exports.

The GST will be a dual tax with levy by both Central and State tax administrations on the same base. The GST demands a well-designed and robust IT system for realizing its potential in reforming indirect taxation in India. The IT system for GST would be a unique project, which will integrate the Central and State tax administrations.

To realise IT implementation, Empowered Group on IT for GST (EG-IT) was setup as one of the governance bodies. 

[Note: The other imp. governance body for GST is Empowered Committee of State Finance Ministers (EC) which Shri Sushil Modi the Deputy CM of Bihar heads and this body was in news recently for several reasons]

EGIT recommended setting up of GSTN SPV as an NIU institution. 

Structure of GSTN SPV:

[Note: Please check my other previous post on SPV in this same thread, to know more about SPV]

Setup as Not for Profit (Section 25) Company. 

So, GSTN was created with following objectives:

1. Simplicity for taxpayers: The process of filing of tax returns and payment of tax should be simple and uniform. 

2. Respect autonomy of states: The design of the IT system should respect the constitutional autonomy of the states. 
[Note: Major questions were raised by the empowered group of state fin ministers on dilution of state's role etc.]

3. Uniformity of policy administration: The business processes surrounding GST need to be standardized. Uniformity of policy administration across states and centre will lead to a better taxpayer experience, and cut down costs of compliance as well as tax administration.

4. Enable digitization and automation of the whole chain: All the business processes surrounding GST should be automated to the extent possible, and all documents processed electronically. 

5. Reduce leakages: A fully electronic GST can dramatically increase tax collections by reducing leakages. 

6. Leverage existing investments: Existing IT investments of states will be leveraged. 

Stakeholders involved in GSTN are:

What will GSTN Do?

GSTN will :

1. Provide common infrastructure and services to Central and State Govts

2. Ensure integration of the Common GST Portal with existing tax administration systems of Central and State Govt

3. Build efficient and convenient interfaces with tax payers and tax administrators

4. Facilitate, implement and set standards for providing common GST services to the Central and State Govt

5. Carry out research, study global best practices and provide training to the stakeholders

IT Architecture of the GSTN Solution for GST:

GSTN will be initially incubated within NSDL - National Securities Depository Limited
NSDL will do all IT development for GSTN. 

Three priority workflow processes of GST administration were identified. These will be implemented as a first phase of roll out of GSTN:

1. Registration: A unique ID is necessary to identify each taxpayer. PAN ID of tax payers will be used to uniqly identify each tax payer. 

2. Returns: taxpayers to file periodic returns to assess whether the taxpayers have computed, collected, and deposited their taxes correctly

3. Challans: Challans are the payment instruments used by taxpayers to actually pay their taxes. Challans are deposited at collecting banks and are forwarded by them to the tax administrations.

4. IGST: Under GST, inter-state trade will be leviable to IGST. Under IGST, the tax paid by the selling dealer in the exporting state will be available as ITC(Input Tax Credit) to the purchasing dealer in the importing state. This requires verification of ITC claims and transfer of funds from one state to another. 

These workflows will be the first set of processes for which IT solution will be developed in the 1st phase of implementation by NSDL. 

See below what other workflows of GST system will be digitised. 

So, the IT architechture or you can say IT Technology Stack (complex sounding IT jargons .. ) of GSTN is:

1. The Common GST Portal links the taxpayer with the varied systems of the tax authorities.

(i) The Taxpayer files taxes through a standardized taxpayer interface - Taxpayer Interface

(ii) States and CBEC implement tax administration systems for assessments, audits, and enforcement within their domain - Tax Administration System

2. Policy decisions are captured in GST Business Rules Engine that defines the tax rates, revenue sharing rules, and exceptions for all parties.
[Note: Engines are servers which posses business rules which is applied to any transaction in that IT system is called Business Engine ]