Coronavirus (COVID-19) means less investment in the firms we rely on for innovative products and services
To the litany of offences on COVID-19’s charge sheet, we will have to add another: a likely downturn in early-stage innovation. This is the argument at the heart of a new working paper I have co-authored with colleagues from NYU, Harvard, and UC San Diego.
This projected downturn is related to patterns of venture capital (VC) investment. It has often been argued that VC funds are inured to the cycle of boom and bust which define public markets, by virtue of being able to take a longer view. By this logic, this would see investment levels in early-stage innovations remain consistent during economic downturns. By the time such innovations came to fruition, we would likely to be on the path towards recovery.
Venture capital funds might be viewed as gatekeepers of innovation
The evidence suggests that this is not the case, however. According to our research, when it comes to early-stage innovation, VC investment is highly pro-cyclical. This means investment levels go up and down with the economy. During recessions, investment in early-stage innovation falls significantly.
In fact, there is evidence to suggest that VC-funding for early-stage innovations falls by an above average proportion, if measured against total investment levels. This would suggest that VC investors react in an exaggeratedly pro-cyclical manner during periods of recession.
This proved consistently to be the case over the four decades of recessions and recoveries covered by the study, from 1976 to 2017. It’s a pattern that can already be observed in 2020. If we compare the five months coming up to March 2020 with the four that followed, we can observe that VC funding of early-stage innovation fell by 34 per cent in the US.
VC funding drives innovation
This has ramifications for innovation. Venture-capital backed firms may account for a small number of firms overall, but they contribute disproportionately to innovation.
If we use citations of patents as our measure, we find that 22 per cent of VC-backed patents were among the top 10 per cent of most-cited patents, with 2.9 per cent in the top percentile, in the period between 1976 and 2017.
VC-backed innovations tend to be of higher quality and economic importance, playing a disproportionate role in job creation and productivity gains.
Venture-capital backed firms contribute disproportionately to innovation
This takes on extra significance in the context of a general slowdown in productivity growth in the developed world. Output per hour is currently growing at 1.4 per cent, half of the levels registered in the 30 years following World War II, or the boom of the 90s and early 00s.
Venture capital funds might be viewed as gatekeepers of innovation. The ideas that they choose to back are often those which go on to have the greatest impact.
The 0.5 per cent of US startups founded each year that receive VC backing go on to account for nearly 50 per cent of those that make it to market. Seven of the world’s top eight firms by market capitalisation (as of May 2020) were VC-backed at some point in their history.
Damage to early-stage innovation
It’s important to note that this decline is very specific to early-stage startups. Later-stage VC investment is not subject to the same effect. Indeed, we see the share and absolute number of VC deals for previously-backed firms increase during recessions. Neither do we see a decline in the quality of innovation for these later-stage firms.
On the other hand, the decline in investment in early-stage innovation during recessions is also matched by a decline in quality. While during boom periods, patents filed by VC-backed early-stage companies are three times as likely to feature in the top percentile of citations, this falls to twice during periods of recession (that they remain higher is a consequence of VC’s gatekeeper function).
We also find that early-stage patents are more limited in scope, drawing on and influencing fewer technological fields during recessions.
More innovative firms are at the greatest risk of losing out
Why does this happen? At least part of it appears to be due to shifts in the types of firm VCs back during recessions. VC funds themselves rely on an inflow of funds from limited partners. Recessions are uncertain times. Even if not faced with an immediate reduction in funds, prudence dictates that they should allow for the possibility of future constraints.
Capital must therefore be reserved for current funding commitments. Many pre-backed firms will have additional requirements during these tougher times. These more advanced firms are also likely to be less capital-intensive, and closer to being cash-flow positive. In other words, safer bets.
Something has to give to make that possible; that thing is funding for early-stage innovations. Research shows that more innovative firms are at the greatest risk of losing out on funding in this context. This consequently has a knock-on effect on the quality of patents filed.
The decline in investment in early-stage innovation during recessions is also matched by a decline in quality
While VC investors may be understandably more cautious during economic downturns, other work has shown the opposite effect during economic booms. In hot markets, VC funds back riskier and more innovative startups. Perhaps we can take some comfort from this.
Yes, we will see VC investment into early-stage innovation fall during the forthcoming recession. But when economic recovery begins to take hold, we can expect to see VC funds once again taking calculated risks on big ideas from young companies. And just because funding may be more difficult to find, doesn’t mean that entrepreneurs will stop having those big ideas in the meantime.
This article draws on findings from the paper “Financial Distancing: How Venture Capital Follows the Economy Down and Curtails Innovation” by Sabrina T. Howell (NYU Stern), Josh Lerner (Harvard Business School), Ramana Nanda (Harvard Business School and Visiting Professor at Imperial College Business School) and Richard R. Townsend (UC San Diego).