AI disruption or opportunity? Why IT services may thrive again

The Indian IT services industry is at a crossroads again, raising questions about how disruptive the next technology wave could be. Interestingly, this is not the first time that the sector has been perceived to be under existential threat; we have been here a few times before.

The first disruption occurred during the 2003-06 period, when SAP started gaining prominence as a one-stop enterprise resource planning (ERP) solution for global enterprises. SAP, with its renewed offerings, offered a plug-and-play ERP solution, which, at that point, threatened the custom application development offerings by Indian IT services companies.

The argument was simple: why would enterprises need IT vendors to design an application for them if a software company was providing the same solution through the purchase of a license? As it turned out, purchasing a license is one thing and customising the application to one’s needs quite another. Eventually, the SAP framework opened up a mammoth ERP segment as an addressable market for IT services companies, and it has remained a large contributor to the industry’s topline even after two decades.

The second disruption happened during the 2015-18 period, when the concept of Software as a Service (SaaS) emerged and cloud adoption was just taking off. We saw the same thesis play out again in 2015. The fear was that SaaS companies such as Salesforce and Workday would replace the existing IT services business model, as enterprises would switch to pay-as-you-go SaaS licenses.

Furthermore, there were concerns about the elimination of work, as SaaS products were easier to implement and maintain. Yet less than a decade later, IT services companies are the biggest purchasers of Salesforce and Workday licenses, as they implement these solutions for their enterprise clients.

The current excitement about artificial intelligence, particularly generative AI, revolves around the ability of these systems to improve themselves or accelerate their own capabilities. Commentators are envisioning a future in which automation and productivity can compound indefinitely at exponential rates.

However, the fact remains that technological diffusion has historically followed an S-curve, with early adoption being slow and expensive. While markets often extrapolate the acceleration phase linearly, history teaches us that the pace of adoption eventually plateaus as organisational integration is costly, regulation emerges and diminishing marginal returns exist in economic deployment. To contextualise this, let’s go back around 10 years, when cloud migration was the buzzword and no one wanted to be left behind. The harsh reality is that even after a decade, estimates for cloud migration vary from about 30% to 50%.

The important thing about software migration is that if an enterprise decides to replace an existing software — for instance, an on-premises CRM application with Salesforce, or Salesforce with some GenAI-based application — they will still need an IT services firm to replace the existing software and integrate the new one into the existing enterprise architecture. The migration exercise requires orchestration, contextual knowledge and domain expertise, all of which are core strengths of IT services firms.

One can argue that it is much easier to use GenAI (or any other new technology) in a greenfield environment, where the entire organisation architecture, data entry, assimilation and interconnectivity of application can be designed from scratch (greenfield) to suit the requirements of the new product or application. However, force-fitting the same into an existing enterprise structure cannot be done overnight, and more importantly, will require significant work around existing applications, thereby necessitating the services of a system integrator, namely IT services companies.

While we are discussing the tech landscape and how it will metamorphose, we should also consider what lies beyond the realms of technology today but may soon come within its ambit because of AI. Consider all the professional services — doctors, lawyers, consultants, investment bankers and journalists — any high-skilled, domain-intensive job. In each of these professions, about one-third of the tasks are repetitive or administrative and can potentially be automated, thereby expanding the market for IT services.

Net-net, in our assessment, AI applications will eventually lead to more work for IT services firms, as the classic Jevons Paradox will be at play. Simply put, Jevons Paradox occurs when technological progress increases resource efficiency, making it cheaper to use, which paradoxically leads to higher, not lower, overall consumption. Proposed by William Stanley Jevons in 1865, this theory highlights that efficiency gains often trigger increased demand that offsets savings, a key concept for understanding sustainability. In the AI context, as software becomes cheaper, there will be more demand for software and hence more system integration work, which will benefit IT services firms.

In terms of valuations, the IT sector is now trading at a discount to the overall market. On a one-year forward PE multiple, the Nifty IT Index is trading at 18.78 times compared to the Nifty 500, which is trading at 20.78 times. As shown in Exhibit 1, over almost 20 years, the Nifty IT index has traded at an average of 1.13 times the Nifty 500 valuation PE multiple but is currently around 0.90 times. Such valuations have broadly occurred twice in the past: during the time of the GFC and more recently in 2017, when the IT services model was expected to be disrupted by cloud and SaaS platforms. The interesting thing is that each time, the Nifty IT Index traded at a multiple of 0.90 relative to the Nifty 500 on a PE basis, the subsequent three-year and five-year returns were very strong, as shown in Exhibits 2 and 3.

In conclusion, while there are dark clouds over the IT services sector and investors are questioning whether the business models of companies in this sector will remain relevant, we have been here before. The sector has not only survived but thrived each time a disruptive technology has emerged. The pessimism around the sector has clearly led to a sell-off, making sector valuations very attractive. Does this present an opportunity for long-term investors? We believe so.