The CMIE database, which tracks project starts and completions, shows capital formation continues to remain sluggish. This ties in with what is happening on the ground. Even as the pace of implementation is gathering speed, the number of projects being completed is slowing down, and in the September quarter, projects worth only Rs 74,500 crore were commissioned. An analysis on investments by Credit Suisse\u2014based on CSO data\u2014shows that, as a ratio of GDP, gross fixed capital formation (GFCF) may have bottomed out, having fallen from 2012 to 2017. During this period, however, contrary to perception, private companies grew GFCF at a fairly brisk 14% compounded at an above-GDP growth rate, with only the metals and the power sectors slowing. This growth was given a big push by the creation of Intellectual Property (IP), a big contributor growing at 20% between FY12-17. The increase in capex via the IP route (primarily software) is not something that was given prominence in earlier data series. However, given it contributes significantly to improving productivity across sectors, it is a welcome addition to the way GFCF is measured. Indeed, in an economy where value addition takes place not just through adding to plant and machinery but through higher productivity, the investment in IP is becoming increasingly critical. While IP may not necessarily create as many jobs as a new cement or steel plant, it would undoubtedly create a lot of value for the economy.Capex in the next couple of years could remain subdued for a variety of reasons. For one, while business houses like the Tatas and the Birlas are investing serious amounts of capital, much of this is taking place through the purchase of stressed assets either via the IBC route or via direct M&A activity. However, this will more likely not show up as GFCF for the year. The reason is that, when the assets are being bought at a discount, it is the negative amount that is factored in the GFCF for the year. Nonetheless, these plants will likely operate at a higher capacity utilisation after they have been taken over by the new owner and this will result in higher output through a lower GFCF\u2014in other words productivity of capital will grow and, more critically, thousands of jobs will be protected. In economic jargon, the ICOR will fall. However, private capex could take a while to get back to the peak seen in FY08. The CS analysis also shows that public sector capex hasn\u2019t gathered speed since 2017 and continues to grow at 11%. However, the big drag between FY12-17 came, not too surprisingly, from household GFCF\u2014including informal enterprises. This segment clocked in a near zero rate of growth, primarily due to weak housing. This seems plausible especially given how inventories of homes have been piling up in the last several years and could take at least four years to be cleared. The other interesting nugget that CS has pulled out of the CSO data is that, between FY12 and FY17, most of the decline in the GFCF growth was in dwellings and buildings and their contribution would have been less than half of what it would have, if adjusted for inflation. With government finances crimped by fiscal compulsions, capex could be subdued for the next couple of quarters. Unless, of course, it manages to find some resources in RBI.