The Department of Science and Technology (DST) recently released Research and Development Statistics (R&D) Statistics 2025-26, the latest edition of its biennial report on the state of India’s science, technology, and innovation (STI) sector. This latest report covers the immediate post-Covid years.Its findings stand in stark contrast to all of the DST’s previous reports: going by its figures, the country’s STI sector recorded its best performance ever in the period 2021-22 to 2023-24. Apparently, gross expenditure on research and development (GERD) rose steeply in this period, making it the most propitious for the country’s STI ecosystem.Chart 1: R&D Expenditure and R&D as % of GDPSource: Research & Development Statistics At A Glance, 2025-26, NSTMIS.This comes as a complete departure from the dismal trend recorded between 2008-09 and 2020-21, during which India’s STI sector showed a declining trend as a ratio of GDP. It was the worst performing among the emerging economies, lagging countries like China by a long distance.The latest report says that the Government expects this strong performance to continue even beyond the years for which the data have been presented, though it has not spelt out the basis for this optimism. Besides the unprecedented increase in GERD, R&D Statistics 2025-26 also reports that one of the major output indicators of R&D, namely, patenting by Indian entities, also increased rapidly during this decade.What led to this dramatic improvement in the performance of India STI sector, and that too within a short interval of just two years? This paper analyses the data provided in the R&D Statistics 2025-26 to find the answer to this question. The statistics on India’s GERD were sourced from the National S&T Survey for the triennium 2021-22 to 2023-24.This survey was conducted for the first time after 2018-19.At the outset, we would seek to understand if the two surveys are comparable in their coverage of entities involved in R&D activities. We would then analyse the statistics to try and explain the basis for the apparent improved performance of the sector, covering both the dimensions covered in the report, namely, the R&D spending by the different sectors of the economy and the patenting activity in the country.1. Basis of R&D statistics: National S&T surveysSince 1973-74, the DST has been undertaking biennial National S&T Surveys to collect statistics on R&D activities in India.The latest R&D Statistics 2025-26 is based on the National S&T Survey conducted in 2024-25, which, as mentioned earlier, was the first to be conducted after 2018-19. This Survey collected “data from around 8000 R&D performing entities covering national laboratories, state research institutions, higher education institutions, public & private industries, scientific & industrial research organisations, etc. spread across the country”.The 2018-19 National Survey, on the other hand, “captured the information from around 5000 R&D organisations viz., public sector, private sector, MNCs, higher education, SIROs [Scientific and Industrial Research Organisations] and NGOs spread across the country”. The two S&T Surveys seem to include largely similar R&D units, except that the earlier survey included an unexplained category, namely, ‘NGOs’. The respective coverages of R&D units in the two surveys are summarised in Tables 1 and 2 below.The major difference between the two surveys is the large proportion of categories that have not been identified, almost 48% in the 2018-19 Survey and 65% in the 2024-25 Survey. Some of these excluded units would be higher education institutions, but these institutions cannot be 65% of the total R&D units surveyed in 2024-25. Between the two Surveys, the number of industrial sector units rose slightly, from 2607 to 2836, but the number of units in unexplained categories more than doubled, from 2393 to 5164.Clearly, a new methodology was used to compile R&D statistics from the year 2021-22. It is from the same year that the data show a sudden, unprecedented leap in R&D spending, a reversal of the earlier trend. Whether or not the new methodology is justified, the data gathered using this new methodology cannot be simply compared to the earlier time series of R&D data.All these issues arise from the lack of transparency regarding the National S&T Survey whose details, including the methodology, are not available in public domain. All the Government has made available are 34 tables, from which the progress the country has made during 2021-22 to 2023-24 can be assessed.Besides providing these data on R&D spending, the DST report also provides data on patenting activity in the country, one of the key output indicators of R&D. Like R&D spending, patenting activity too has substantially increased in India over the past few years.2. The broader pictureIn 2023-24, India’s gross expenditure on research and development (GERD) was Rs 2.45 lakh crore, up from Rs 1.95 lakh crore in 2021-22, an increase of nearly 25% in two years. But considerably more significant was the near doubling of India’s GERD from 2018-19. It jumped 53% in a single year, 2021-22. None of the economies for which comparable data are available from the year 2000 have experienced such a steep rise before.Chart 2: Annual growth rate of GERD and GDP at current pricesSource: Research & Development Statistics At A Glance, 2025-26, NSTMIS. The increase in GERD was also reflected in India’s R&D intensity, or the ratio of GERD to gross domestic product (GDP). This registered a steep increase, from 0.64% to 0.84%, between 2020-21 and 2023-24. The DST projections indicate that R&D intensity would have increased to 0.87% and 0.9% in 2024-25 and 2025-26 respectively, levels that India has never seen before. However, unlike the figures for the immediately preceding years that are derived from its National S&T Survey, 2024-25, the DST has not provided the basis for these projected increases in R&D intensity.The rise in R&D intensity from 2021-22 is significant, as its trend decline beginning in 2008-09 was reversed. During the 12 years following 2008-09, India’s R&D fell from its peak of 0.84% of GDP to 0.64% in 2020-21. As a result, India’s rank among the countries for which UNESCO reports R&D intensities fell from 33 in 2009 to 45 in 2020, while during the same period, China improved it R&D intensity from 1.7% to 2.4%, and its rank from 19 to 12.Now the steep rise in R&D intensity post-Covid raises two key questions: one, how was India’s 12-year slump made up in just three years, and two, on what basis has the Government predicted that the upswing will continue unabated? Before we try to find answers to these questions, we shall analyse the patterns of R&D expenditure, as between the public and the private entities.It is not just the increase in R&D intensity that the DST has reported in its latest S&T indicators; the relative contributions of the public and private entities in GERD also changed. Private entities, according to the DST, include private sector industries, Scientific and Industrial Research Organisations (SIROs) and private higher education institutes. Until the end of the previous decade, the share of the publicly funded entities in GERD was consistently above 60%, with the highest level being reached in 2004-05, when these entities contributed over 75% to India’s GERD.Though this figure declined during the period after 2010-11, the public entities’ share was still above 66% in 2019-20. However, during the current decade (2020s), private entities’ share has risen dramatically. DST’s projections show that the private sector’s share in GERD would have been almost 60% of India’s GERD in 2025-26 (Chart 3).Chart 3: Share of Public and Private Entities in India’s Gross Expenditure on R&D (GERD)Source: National Science and Technology Management Information System, S&T Indicators TablesThis increase in the share of private entities in GERD was largely due to the increased spending by private sector industry, which increased its spending on R&D by 43%. Increased R&D spending by private sector industry also improved its share in Business Enterprise Research and Development (BERD), which includes public sector industry as well (Chart 4).Chart 4: Share of public and private sector industry in India’s Business Enterprise on R&D (BERD)Source: National Science and Technology Management Information System, S&T Indicators Tables3. A disaggregated account of R&D spendingThe contribution of key segments to the rapid increase in GERD during 2021-22 to 2022-23 can be best understood if these figures are juxtaposed against those for the immediately preceding triennium (2018-19 to 2020-21), during which India’s R&D intensity had fallen to its lowest level. In the latest triennium for which DST has provided detailed data, though all segments of the STI sector contributed to the sharp increase in India’s GERD, two segments, namely, the higher education sector and private sector industry, were the best performers (details in Annex Table 1). Between 2021-22 and 2023-24, the higher education sector’s R&D spending increased by over 28% while private sector industry’s spending increased by nearly 43%. What is the basis for these increased expenditures?The increase in higher education institutions’ spending on R&D was driven by the publicly funded institutions, which accounted for almost two-thirds of the total spending in this sector during 2021-22 and 2023-24. Although disaggregated data for the higher education are not available for the previous triennium, it may be prudent to assume that the share of publicly funded institutions would have been similar in the closing years of the previous decade.But the significant increase in R&D spending by publicly funded institutions in higher education appears counter-intuitive, given the decline in the share of higher education in the Central Government’s budgetary spending, falling from 1.4% in 2014-15 to 0.83% in 2023-24.More importantly, the share of higher education in budgetary spending remained consistently below 1% in the post-Covid period. Thus, during a phase in which the Central Government was spending less on higher educational institutions, is it plausible that these institutions were spending more on R&D?As stated earlier, it appears from the report that private entities have emerged as the main drivers of India’s STI sector, with a share of almost 52% of GERD in 2023-24. We had also mentioned that DST had explained that the data it had presented were obtained from the National S&T Survey, 2024-25, which collected data from around 8,000 R&D performing entities, including from the private sector. The survey included private sector entities, corporate entities, scientific and industrial research organisations (SIROs), and higher education institutions. Over 87% of private sector’s R&D spending came from corporate entities and, therefore, these need closer scrutiny.DST’s Survey obtained data for the private sector industry for the latest triennium from over 2700 units across 41 industries. Aggregate spending on R&D by these units was Rs 1.1 lakh crore in 2023-24, increasing from about Rs 77,000 crore in 2021-22.Two aspects of R&D spending by the private sector industry need emphasising. The first is that R&D spending undertaken by the private sector units increased by 43% in two years. Among the large spenders on R&D, biotechnology, transportation, and information technology, increased their spending by more than 50%. The second, and more significant, aspect was that R&D activities were highly concentrated in the top 10 spenders; they accounted for 93% of the total spending (Annex Table 2).Such extraordinary increases in India’s recent R&D spending raise several questions, especially when such spending had stagnated for well over a decade until 2020-21, a point we had made at the outset. Questions regarding the integrity of the R&D data have become even more glaring due to the absence of granular details regarding National S&T Survey, 2024-25 and, consequently, the lack of transparency regarding the sources of R&D data.Equally important is the fact that the DST has not defined some of the categories of entities it has included in the Survey. We provide two examples to elaborate this point.First, we compare DST’s data with those of the Prowess database of the Centre for Monitoring the Indian Economy (CMIE). The latter covers large public and private limited companies, who conduct most of the R&D expenditure undertaken by private sector industry. We consider the two largest R&D spenders, according to the DST Survey, namely, the automobile and pharmaceutical industries.DST reports that R&D spending undertaken by 142 units in the automobile industry was Rs 39,137 crore in 2023-24, which was more than three times the R&D spending of 172 companies figuring in the Prowess database. In the case of pharmaceuticals, the difference in R&D spending between the two data sources was not as large: 492 pharmaceutical units included in the DST Survey spent about Rs 26,000 crore on R&D in 2023-24, about a third higher than the pharma companies in in the Prowess database. The crucial difference between the two sets of estimates is that data obtained from Prowess are verifiable, as they are from the companies’ annual reports, but DST’s data cannot pass this test, as the data of the National S&T Survey, 2024-25 are not available in public domain.There is no valid justification for keeping the details of the S&T Survey data out of the public domain, since placing them in the public domain would have dispelled any doubts regarding its quality, such as are being expressed now. Survey responses could be kept confidential, as is done in the case of other data sets, such as the Periodic Labour Force Survey.The lack of definition is most glaring in DST’s reporting of R&D spending by “multinational enterprises” in India. As a result, the basis of this set of data too is questionable. The latest survey provides data pertaining to 390 such “multinational enterprises”, as against 382 enterprises included in the survey for the previous triennium, namely, 2018-19 to 2020-21 (Annex Table 3).The definition of multinational enterprises adopted here has not been provided, which is a major aberration.The veracity of the numbers on R&D by multinationals is doubtful, as India’s private sector multinationals (mentioned in Annex Table 2 as “private sector – India”) are reported to have increased their R&D expenditure three-fold in a single year, between 2020-21 and 2021-22, while the number of companies included in this category grew by just 19. Further, BERD undertaken by the surveyed companies increased by almost 50% during the triennium covered by the National S&T Survey, 2025-26. It is very doubtful that R&D spending can grow by this magnitude in just one year.The DST claims that the newfound dynamism in India’s STI sector is not only restricted to vastly improved R&D spending, but it also extends to one of the key output indicators of R&D, namely, patenting activities. We discuss this aspect in the next section.4. Increase in patenting activity in India: What lies behind the numbers?Patenting activity in India, both in terms of the patent applications filed and sealed, has gone up significantly, almost coinciding with the spurt in R&D spending reported by the DST (Annex Table 4).In 2023-24, patents granted in India crossed the 100,000-mark for the first time. Perhaps more significantly, patents granted in the country reached the five-figure mark only in 2017-18, but in the following five years, they grew eight-fold. Such increased patenting activity may not as such be a positive development, for the reasons elucidated below.First, even if we accept the argument that a mere increase in patent grants is a positive augury, it cannot be assumed that India would benefit automatically, as patents are granted in India to both residents and non-residents. If the larger share of patents is granted to the non-residents (foreigners), which is indeed the case, the benefits of higher levels of patenting do not accrue to the country. Foreign patentees zealously protect their patents and do not usually allow their use in India.Secondly, cross-country evidence suggests that mere increase in patents granted to residents is not a robust indicator of a country’s innovation system, and hence of its improved technological competence. The quality of the patents granted holds the key, an issue which has frequently been discussed in the context of the rapid increase in patents granted in China.Who owns the larger share of patents granted in India? Data for answering this question are provided by the World Intellectual Property Organisation (WIPO). Using these data, we can not only understand the trends in the relative shares of patents owned by residents and non-residents in India, but can also compare the numbers for India with those of three other countries, namely the US, Japan, and the Republic of Korea (RoK), all of which grant patents in excess of 100,000 annually.The share of residents in the patents granted in India increased over the past decade, from nearly 11% in 2014-15 to 32% in 2024-25. In absolute numbers, patents granted to residents increased from less than 700 to almost 10,700, while the corresponding figures for non-residents were 5,200 and almost 23,000, respectively.Cross-country comparison shows that the share of patents granted to residents in India was the lowest among the countries that granted 100,000 patents in 2024. Take the case of China, which was the leading country in terms of patent grants for several years: in 2024, 90% of the patents granted in China belonged to its residents. In both Japan and the RoK, more than 75% of the patents granted were owned by residents over the past decade, while the corresponding figure for the US was below 46%.What are the implications of the aforementioned patenting trends for India? Non-residents, which are mostly multinational corporations, have used their patent rights to establish/strengthen their control of India’s technology market. Foreign firms have thus used their monopoly power to pre-empt entry of potential Indian competitors, thereby perpetuating India’s technological dependence.The immediate consequence of this technological dependence is steady increase in India’s liabilities on account of payments for the use of patents and other forms of intellectual property, such as trademarks, over the past decade. Between 2014-15 and 2024-25, foreign exchange outgo on this account increased more than three and a half times, from $5 billion to over $17 billion.That patent numbers are inadequate indicators of the state of a country’s innovation system has widely been discussed in the context of the extraordinary surge in patents granted by the China National Intellectual Property Administration (CNIPA). It has been argued that the quality of patents granted in China was “below the world average and significantly lower than that of other developed economies”, which explains the sudden surge in grants. Even the CNIPA took cognizance of the growing criticism by shifting its focus from quantity to the quality of patents that it grants.The US Federal Trade Commission (FTC), the competition authority in the US, had warned about the dangers of granting poor quality patents. The FTC argued that patents of questionable quality “may have anticompetitive effects, can cause unwarranted market power, and can unjustifiably increase costs. Such effects can hamper competition that otherwise would stimulate innovation”.Given these concerns about patent quality, questions will be asked about the patents granted by Indian authorities in 2023-24, for two reasons. First, the Patent Office granted 250 patents every working day that year, an ‘achievement’ that raises serious questions, as the institution is known to suffer from shortage of “manpower, physical and digital infrastructure”. This issue has been underlined by, among others, the PM’s Economic Advisory Council.Secondly, since patents are an output indicator of R&D activities, steep increase in patenting activity should have resulted from the growing robustness of India’s innovation system. Unfortunately, even after decades of discussions on ways to strengthen the country’s technological sinews through increasing R&D intensity, India continues to considerably lag behind most major economies and remains a technologically dependent country.The evidence in this regard is the steep rise increase in India’s payments for the use of intellectual property, due to its increased use of imported know how and trademarks. These payments have risen 2.5 times since 2020-21, and quadrupled overthe past decade (see Table 3).Table 3: India: Charges for the Use of Intellectual Property ($ bn.)ConclusionsTo sum up,The steep rise in R&D spending claimed in the latest official publication by the DST raises many doubts. It is based on a new survey methodology; however the methodology and detailed data have not been placed in the public domain.These data show a jump in R& D spending of over 50% in a single year (2021-22), a near-doubling in five years, and a complete reversal of the earlier declining trend. That declining trend had been an embarrassment for the Government.In just three years, R&D intensity (i.e., gross expenditure on R&D as a ratio of GDP) reversed the decline of the previous 12 years.Private corporate sector’s spending on R&D was the main driver of this growth, rising so sharply that private spending on R&D became the dominant share, reversing the earlier respective shares of public and private R&D spending.Our doubts are strengthened when we consult other data sources. The figures of private corporate R&D spending reported by the CMIE Prowess database are far lower than those reported by the DST. While the CMIE data are independently verifiable, the DST data are not in the public domain.Even as the share of higher education in central government expenditure was falling, the report shows R&D spending by publicly funded higher education institutions rising. This too seems doubtful.While patents granted in India grew eight-fold between 2017-18 and 2023-24, more than two-thirds of these were granted to non-residents, which would perpetuate the control of multinational corporations on India’s technology market.The quality of patents granted is open to doubt, as India’s Patent Office is woefully understaffed and under-endowed.Republished with permission from Research Unit for Political Economy. Read the original post.Biswajit Dhar is former professor of economics, Jawaharlal Nehru University.