Print This Post

Despite huge investments by the government and social enterprises, India’s education sector continues to be a story of poor learning outcomes and inadequate quality. What lessons can we learn from our experiments so far?

Numbers related to India’s education sector are mind-boggling, to say the least. With more than 1.5 million schools (1.1 million of them run by the government) and more than 250 million students enrolled in them, the country’s K-12 school system is one of the largest in the world.

India’s education market, currently valued at USD 100 billion , is expected to nearly double to USD 180 billion by 2020, fueled by a growth of 30 percent CAGR in the online and digital learning market over the same period.
There is a significant mismatch between education spending trends and learning outcomes.

In terms of government spending, the Union Budget has pegged an outlay of INR 79,685.95 crore for the education sector for 2017-18. Of this, nearly 60 percent (INR 46,356.25 crore) is for the school sector and the rest for higher education.

On the corporate front too, education received the highest amount of CSR funding among all social development activities in 2015-16: 920 National Stock Exchange-listed companies together spent INR 2,042 crore on education, up 30 percent from INR 1,570 crore in FY15.

Add PE money, strategic investments, and grants provided by various foundations–and the amount of money in circulation in the education space is quite substantial.

Despite these large investments by both the government and private sector, quality remains unsatisfactory. There is a significant mismatch between education spending trends and learning outcomes that calls for a serious introspection of our education policy and practice.

Where did the money go?

Private capital
The Indian education landscape can broadly be classified into pre-school, K-12, higher education, tutoring and test prep, vocational training, and multimedia & ICT.

According to VCCEdge, of the 289 deals worth USD 919 million registered since 2012, more than 56 percent has gone into test prep (USD 217.57 million) and e-learning portals (USD 301.29 million).
Within the test prep segment, learning app maker BYJU’s alone received USD 150 million in 2016-17. This was approximately 50 percent of the overall VC money received by all ed-tech companies till December that year and 22 percent of the total money received in the K-12 segment in the last seven years.

That’s not all. Even within the K-12 segment, approximately 50 percent of the investments have gone to test prep and e-learning.
The table below highlights the distribution of the major equity deals in this segment in the last six years.

India's education sector

India's education sector

This disproportionate investment in certain segments has resulted in some imbalances in the sector:

  • New entrepreneurs generally see this as a signal that the segment is lucrative and develop sub-par business models that are not aligned with the objective of improving learning outcomes.
  • Public and private investment in earlier stage education (pre-school and K-12) is proving to be insufficient resulting in a large gap in the skills taught and those required by industry. This disparity between the knowledge acquired in school and those evaluated through standardised tests has also resulted in a burgeoning test-prep industry.

This, in turn, has diverted funds from the other segments into this lucrative category, setting in motion another cycle of skewed investment patterns.

Government spending
The anomalies in investment are reflected in government allocations as well, but of a different kind. As against an allocation of INR 5 crore last year, the school assessment programme has been allocated a paltry INR 67 lakh in the 2017-18 budget. Conducting learning assessment in a school system comprising 250 million students in more than 1.1 million schools with this amount is a tough ask, especially when all the players acknowledge that learning outcomes are critical.

On the other hand, the mid-day meal programme has been allocated INR 10,000 crore, up by INR 300 crore from the last budget. This meagre increase is unlikely to create a difference in the functioning of the scheme. However had this INR 300 crore been allocated to the Digital India E-learning initiative for higher education, which has been allocated INR 497 crore, or for the Department of School Education and Literacy (which has a meagre INR 14 lakh budget), both programmes would have benefited greatly.

India's education sector
Photo Courtesy: Charlotte Anderson Photography

Despite all these efforts, why haven’t we got it right?

Such skewed investment patterns point to some underlying problems in the country’s education sector and the inadequate response to them from various stakeholders, including government, corporates, and social enterprises.

Are we solving the wrong problem?
The government went in with the assumption that all teachers were good at teaching and would enhance learning outcomes, regardless of the context they taught in. This assumption allowed them to limit their role to providing access to schools for all children (through RTE) without paying any attention to the existing pedagogical system. That assumption has fallen flat.

Although ‘learning outcomes’ has become a talking point for policy makers, the government has chosen to not act on it even after admitting the flaw in their theory. The meagre budgetary allocations to school assessments and digital literacy are reflective of the government’s lack of focus on educational outcomes.

Further, a break-up of government spending shows that only 0.8 percent goes towards capital expenditure, while 80 percent goes towards teachers’ salaries, leaving little to be spent on infrastructure creation, which eventually translates into ineffective infrastructure and poor quality of education.

Technology isn’t the only answer
Several new-age entrepreneurs in the education sector believed that technology alone would improve the quality of education. Private players, entrepreneurs and tech believers assumed that well-designed products focussed only on enhancing the learning outcomes of students could replace teachers in the classroom and solve the problem of poor teacher quality.

This hypothesis led to the rise of business models and products with an inadequate focus on classroom execution and facilitation, leading to the failure of many good ideas that could have been successful if only the entrepreneurs had been careful about this assumption.

Over-regulation
The highly-regulated environment in the education space is a barrier to attracting quality entrepreneurial talent to K-12 education as compared to the higher education segment. The rise of innovative solutions in test prep and digital education can, to an extent, be attributed to this gap.

A blinkered view
There is little comprehension of the fact that the education sector’s challenges cannot be solved by attacking individual problems: it needs a concerted effort where problems are addressed simultaneously. Thus, teacher absenteeism must be tackled along with the problem of poor quality of teacher training. The problem of test prep must be solved along with the provision of good digital literacy to students in schools, especially in rural and semi-urban areas.

The consequences of not following a multi-pronged approach can be detrimental to the sector, as we realise the distorted impact of such moves only after a few academic cycles.
This is because learning outcomes are dependent on the exam cycle of schools and colleges. Any business model being experimented within the K-12 segment therefore takes at least two academic cycles to get validated in terms of its efficacy in creating impact.
An acceptance of this fact might help the entrepreneurs and investors become more patient and stick to fundamentally better solutions than just looking for easier exits.

India's education sector
Photo Courtesy: Charlotte Anderson Photography

What can the sector do differently?

The education system cannot be transformed by working in silos. Only when we take a holistic perspective can we talk about how impact investment is helping create the desired change in the sector.

Investors, foundations, governments, educationists, activists, entrepreneurs and other stakeholders are investing considerable efforts into the sector. But we need more. Here are a few things to consider:

  • The incentives for all stakeholders–parents, teachers, students, principals, and school owners–should be understood with the goal of aligning them as far as possible to ensure improvement in learning outcomes. There is a great deal of literature available on the importance of incentive alignment and contract design, and it is time we translated these learnings into practice.
  • Impact measurement metrics should be clearly defined and integrity of the data collected from the field should be validated. Pockets of excellence and average numbers on certain chosen parameters are not truly representative of the state of education in India.
  • Weak business models that do not have well-defined impact or where the costs are too high compared to the impact generated should be allowed to perish. It is important for the market to let bad business models fail.
  • Despite being the second largest education system in the world, there is a dearth of quality faculty at scale. There is a huge need for both quality pre-service and in-service training for teachers.
  • Entrepreneurs should recognise that teachers are pivotal to success and must include them while designing the product or service. If there is no value proposition for them in using the product, then product adoption is going to be low and ineffective.
  • Corporates should go beyond supporting tuitions, providing books, or building school infrastructure with their CSR funds, says Prachi Jain Windlass, director at the Michael and Susan Dell Foundation. “Firms could look to support processes organisations and initiaitves that have clear learning outcome targets, which are also easily measurable,” she says.

When we ask the right questions, identify the issues that really matter, and find and implement the solutions that address these issues, our children might learn better–and the numbers might begin to add up.

Research for this article was partly taken from ‘Impact Investing in K-12 Education in India‘ by Sudhanshu Malani (Investment Associate), Villgro Innovations Foundation.

Vikash Jha

Vikash Jha

Vikash is the Investment Associate for Education at Villgro Innovations Foundation. He has over five years of work experience in the for-profit and nonprofit sectors. He was a Teach for India fellow, prior to which he worked with Tata Motors in their Pricing and Supply chain department. Vikash also has entrepreneurial experience with Just For Kicks, which he helped set up. He holds an MBA from Indian School of Business, Hyderabad.

9 Comments

  1. Geetha Manoneeta Reply

    My comment is for both Vikash and Ankit.

    Ankit !…. I DO NOT …at all agree that ….”There is not enough scope for innovation in teacher training for a private sector company. Teacher training is not something that you can show to the parents and the willingness of the school to pay is very low”……..
    I have an innovative, multi-pronged idea to address this issue and I am presently discussing this with Govt./Schools…..I am confident of making my model successful, as I will have the backing of a robust network of quality teacher-trainers for working teachers.

    I TOTALLY agree with you that school children are obsessed with EXAMS …rather than enjoying the experience of LEARNING. …

    I am really SURPRISED that there is very little mention about the merits of OLD systems prevalent when I went to school. At that time…LEARNING was the focus, Languages were given due importance, which helped me grow as a sensitive and sensible human being. character, discipline and virtuous conduct was more important to my parents and teachers than my marks card….. Is is not something that needs to be nurtured to create a better society ? …should we not replace some of the buzzwords like COMPETITION with Self-competition is the best competition..that leads to excellence ? Why everything has to be measured in terms of material gains only ?

    The best RETURN or REWARD to society by the education system is giving the society BETTER INDIVIDUALS, BETTER CITIZENS who have a caring attitude towards the people around are willing to lend a helping hand to those who need their help.
    Geetha Manoneeta, Bangalore

    The BEST solutions will emerge ONLY when WE ALL AGREE on the SOLE purpose of EDUCATION…….Preparation fora dignified and honorable life…..( and not just eking out a living or chasing money all the time somehow or the other)

  2. Pingback: The numbers don’t add up | Villgro

  3. Completely agree with the above point. When we are trying to look at core purpose as improvement of learning outcomes the focus leads to standardization and as seen in America through last two decades teaching for the test. Emphasis by the govt needs to be empowerment of govt schools and pushing state govts to set up model schools especially in the urban settings. Pvt schools especially affordable private schools have become a huge market for the service providers and even lot of investors are pushing new start ups to design solutions for APS. Almost all of these programs dont exceed beyond one or two years in the school which is not enough to set up quality practices thus eventually disappointing students and teachers.

  4. Dear IDR moderator, why did you decide not to publish my previous comment? Sounds contrary to the culture of debate you are trying to foster at IDR… Happy to hear the reasons

    • Devanshi Vaid
      Devanshi Vaid Reply

      Hi Clementine,

      We did in fact publish your comment (and love what a detailed and thoughtful one you wrote). However, there was a delay in publishing it only because it was in the middle of the night in India, and we are based out of Mumbai.

      Best,
      Devanshi (on behalf of IDR)

  5. As an entrepreneur in this space, I would like to comment on the argument that one should take a holistic perspective on the problem. As much as I agree with that I would like to point out that all the incentives in the funding space are skewed AGAINST that. If you want to get through the investment jungle, you need to have ONE PROBLEM, that can be described in 100 words, that you tackle. There is very little space for complexity and holistic perspectives. It’s not surprising at all that most entrepreneurs end up focusing on one aspect of the problem and forgetting the rest. At Slate2Learn we are stubbornly trying to keep that holistic approach to solving education technology problems, but it means we need to constantly paddle against the tide.

  6. Vikas, Thanks for writing this.

    I have a different view on what you have mentioned about private investments. If entrepreneurs want to succeed in the education sector they will have to develop products that the market needs. Not what you and I believe that is best for the education sector. The statement that because entrepreneurs have developed products that are focused on test prep, and that is why investors have decided to ignore other segments is far from the truth. Investors want a return on their capital and they will always invest in market led solutions. Again not what is the best for the education sector as a whole.

    Regulation is definitely not the reason why startups are focusing on test prep. The only objective of the students is to pass the exam, to get into a good college, to get a job and ultimately to lead a better quality life. The premise that the student wants to learn is a false premise and any innovation towards digital learning is to help the student for the first step which is passing the exam.

    Your point about entrepreneurs needing to work with teachers also is problematic. Teachers really don’t have a say in how and what a student should learn. There have not been any startups that have succeeded because they worked with or for teachers. The products are built for the paying customer which in this case are the parents. I can’t name a single company that has scaled by forcing the teachers to use their solution. There is not enough scope for innovation in teacher training for a private sector company. Teacher training is not something that you can show to the parents and the willingness of the school to pay is very low.

    I would like to point out several other points of your article as well but the bottom line is that the private sector’s end goal is not to ‘improve’ the education system and ‘improve’ learning outcomes for each child but the first objective is profit for the stakeholders involved. This point the article fails to understand which has led to many incorrect conclusions.

    • Vikash Jha Reply

      Hi Ankit,

      Thanks for the engaging with the article. I do value your inputs and I am delighted to see a diverse set of views existing in our ecosystem on multiple issues. It only augurs well for the future. Some of the points which you have raised I have tried to address on LinkedIn. Will be happy to talk in detail over a call or a meal sometime.

  7. There are few important points being made in the article about teacher quality and the need for more integrated approaches. What is not clear is the author’s assumption about e-learning and assessments as being the core solutions. Also to be led by learning outcome measurement as the core hugely narrows the scope of education and its purpose.

Write A Comment

Shares
Close Icon