Recent studies simply emphasize the continued failure to develop adequate infrastructure to complement and support innovation at the level of individual farmers. These studies document numerous deficiencies in hard and soft infrastructure for alternative crops. For kinnows, one study notes the lack of adequate facilities for storage, grading and waxing, and for refrigerated transport. A case study of failures in processing infrastructure highlights the government’s complete dysfunction in building and operating juicing plants: while Pepsi in this case could just shift to alternative locations, local farmers were left in the lurch, and the government’s large investment was mostly wasted. Again, there is an undercurrent of government failure tied to political economy factors in this story. There are yet other inadequacies in the case of infrastructure for kinnows, including lack of local aggregation facilities for different grades and types of crops, inefficiencies associated with intermediaries, and lack of access for smaller farmers. In this context, Govil makes the point that even larger farmers who are trying to diversify into non-food grain crops can feel poorly treated when they participate in new value chains, since they bring in relatively small quantities for sale. Inadequate drying, treatment, and processing facilities, inefficient and uncompetitive intermediation, and inadequate storage and transport facilities all appear as deficiencies in the infrastructure,growing blueberries for crops as diverse as maize, moong and turmeric .

Lack of adequate provision of inputs and technical advice, already discussed in the previous section, are another deficiency in the supply chains for multiple alternative crops. A second Munjal Institute for Global Manufacturing study documents similar deficiencies in relative success stories such as wheat and dairy production, as well as for a range of vegetables. An additional deficiency brought out in this study is the asymmetries in information that pervade the value chain, contributing to coordination failures, greater uncertainty for farmers, price volatility, and inefficiencies in market clearing. The Munjal Institute for Global Manufacturing study discusses information technology solutions as ways of overcoming some of the pervasive inefficiencies, but these solutions also require significant up-front investments. To summarize, Punjab has not developed the physical and institutional infrastructure required for crop choice innovation and diversification. Individual farmers cannot create this infrastructure, which requires large economic actors – typically a mix of public and private – to participate. Connecting to the theme of the last section, the infrastructure needed is also more complex, sophisticated and expensive than what was created in the 1950s and 1960s, at that time largely by the state government, but with the underpinning of a national food security policy .Switching costs were first analyzed systematically by Klemperer , in the context of imperfect competition. The term includes learning costs, transaction costs, physical investment costs and artificial costs created by oligopolistic firms, such as rewards for loyalty. In all these cases, the focus is on consumer switching costs, although the customers may also be firms that are purchasing equipment or other inputs from supplier firms.

The latter case is most relevant for considering the case of farmers as potential innovators. Beggs and Salies consider innovation in the presence of switching costs, but their focus is on the incentives of sellers to innovate, rather than of buyers to adopt innovations – although the latter affects the former. In the context of farmers in a state such as Punjab, there are challenges for the providers of technology as well as for the potential adopters , and furthermore for the buyers of the farmers’ output. As discussed in the last two sections, there appears to be insufficient investment in various aspects of the agricultural value chain, both upstream and downstream from the core production process of growing the crops. To the extent that farmers face switching costs, such as those associated with learning new production techniques as well as those associated with learning how to market and sell new kinds of produce , this can inhibit adoption of innovation by farmers. That, in turn, can reduce investment by suppliers of farmers and their buyers, and do so in ways that lead to sub-optimal outcomes. The idea here is that switching costs interact with market imperfections to create inefficiencies. It is useful to clarify here that switching costs are one-time costs, whereas many of the challenges discussed in the previous two sections involve ongoing costs associated with alternative crops, such as cold chain infrastructure, which will be costlier to build and maintain than the existing infrastructure for food grains. The uncertainties associated with new technologies and new crops are one source of switching costs, since these uncertainties attenuate and ultimately disappear with experience and learning. However, there are also ongoing risks associated with alternative crops. Some of these risks are inherent in the differential susceptibility and response of these crops to weather variation, insects and disease. Other risks are a product of government policies. In particular, it is well-known that the Public Distribution System and associated Minimum Support Price reduce price risks for farmers, and bias their production choices towards wheat and rice.

However, removing price risk does not remove production risk, and a potentially better alternative would be some kind of crop insurance, which addresses production risk, and which could be extended to a range of different crops. For example crop insurance tied to publicly observable rainfall levels can be customized for different crop choices by adjusting contract terms. Such insurance can also potentially reduce the risks associated with switching to new crops, by adjusting premia based on experience, perhaps with government subsidies . In the case of Punjab farmers, some of the problems associated with crop insurance experiments in other parts of India, including unaffordable premia and high transaction costs, would still be present, but might actually be less significant, given the higher income levels in Punjab. What is important to recognize is that there is enough specificity and sufficient market size to create crop insurance products at the state or regional level,square plant pots and Punjab seems to be behind the curve in this respect . States such as Andhra Pradesh and Gujarat seem to be among the leading sites for crop insurance experiments, but the Punjab government could pilot insurance schemes for alternative crops rather than relying on the national government as it seems to be doing . Experience from other states can help in this respect. The Green Revolution involved the implementation of what might be termed frontier innovations: the development of new high-yielding varieties of wheat and rice. Complementary innovations were required in terms of fertilizer and water use, cropping patterns, sowing and harvesting techniques, and so on. None of those complementary changes represented fundamental technological or even organizational innovations, in the sense that they represented pushing out the global knowledge frontier. This phenomenon of multiple adaptations or incremental changes surrounding a more basic kind of technological innovation is typical. A good example of this process has been the productivity of different generations of microprocessors, where there is a learning curve for each new major advance in technology, as a result of many small adjustments in the complex manufacturing process. In the case of Punjab agriculture, dominated as it is by wheat and rice, with cotton and sugarcane also being significant crops, the innovation process currently appears to consist of attempts to continue developing new varieties of these crops, through conventional hybridization experiments, as well as genetic modification trials. Some of this innovation is required simply to deal with new diseases, or pests that have become resistant to methods of chemical control. In other cases, it seems that frontier technological innovation is subject to diminishing returns. For innovators, such as scientists at agricultural universities, the kinds of research they pursue may be driven by risk-reward trade offs that favor these directions of potential innovation, despite the problem of diminishing returns. These directions are safer in terms of probabilities of success, and more directly of value since they provide improvements in areas that can benefit a large number of cultivators. This argument has considerable overlap with the ideas in Christensen , in which he argues that successful firms may under-invest in disruptive innovations. One can extend this possibility to include research in other organizations, such as universities or government laboratories. There is an additional aspect of this problem. The previous paragraphs highlighted the incentives of researchers, in terms of where they focus their efforts, with respect to crops and types of innovation.

The status quo will favor incremental innovation in crops that are currently grown most commonly. However, there is the additional problem that research in crops that are not widely grown, whether it leads to incremental or discontinuous innovation, will require a multitude of changes at the farm level, in detailed adaptations of growing techniques, as well as innovations in other parts of the agricultural economy’s infrastructure and value chain, as has been highlighted in earlier sections of this paper. In other words, alternative choices of innovation effort by researchers may have uncertain payoffs, with that uncertainty depending on institutional factors beyond their control. In the Green Revolution case, some of the physical infrastructure had been established in the decade preceding the actual introduction of high-yielding varieties. Yet another problem is that some of the adaptive innovations and adjustments in growing techniques for alternative crops are not rewarded within the scientific establishment. Compared to the Green Revolution era, there is a stronger indigenous scientific base, but it is arguably the case that the level of detailed understanding of complementary adjustments in farming practices that are needed to make a the core technological innovation successful – that is, the range of activities captured under the broad heading of “agricultural extension” – has not progressed in the same way as the core scientific research, and in some ways has even regressed. This issue is a controversial one, of course, and it is not restricted to Punjab, but is India-wide . Furthermore, it is just one example of the deeper problems of delivery of public services – health and education are commonly discussed as examples of public sector failure in India – and public agricultural extension is in some ways doing better than those cases, especially given its unique challenges in the context of implementing complex sets of technological and behavioral innovation. The issues raised in this section are somewhat different than more traditional economic analyses of adjustment costs as barriers to adoption of innovations.The argument here is based on problems with incentives within organizations that produce innovations, and related issues of externalities resulting from the need to have coordinated efforts across different aspects of the production process, in order to have successful innovation. This suggests that the policy response is perforce going to be challenging to implement, since it will require higher-level interventions to overcome externalities and coordination failures. This raises the difficult question of how appropriate incentives for these higher-level policy interventions can be created, and this problem is taken up in the conclusion.The Green Revolution in India was led by the public sector. National priorities for food security shaped policies and institutions for public procurement. The state government played a crucial role by providing physical infrastructure such as market towns and connecting roads. Research and other support for innovation came from public universities, and public sector financial institutions played a role in provision of credit. Finally, the public sector has been heavily involved in the production or provision of key inputs such as water, electric power and fertilizer, typically with significant subsidies. Despite all of this public sector involvement, the role of the private sector in Punjab’s agriculture has obviously not been negligible. This statement includes, but extends beyond, the crucial role played by private sector enterprise, in the form of efforts by individual farmers, large and small. It is difficult to quantify the relative entrepreneurial abilities of Punjabi farmers, but there is some formal analysis of their entrepreneurship , and plenty of indirect evidence of Punjabi farmers’ ability to succeed in widely varying locations, including traditional destinations such as Australia and California , as well as newer ones such as Italy and Madagascar .