The Year in Valuation: It’s “In” to Be Human

Photo by Pixabay via Pexels

smaller data sets, qualitative measures, and human intuition are suddenly in vogue among investors.

Pandemics change the way we act. They change the way we think. And, not surprisingly, economists are recalibrating how investors value the changes in our actions and thoughts. Value, it appears, has become a dangerously-shaped, often political football. Once reliable sources for news and data have been subsumed by their partisan audiences. Those from the political right see risks where those from the left see rewards. Winners and losers are becoming harder to pick. For example, we tracked roughly 480 initial public offerings in 2020. Assuming each equity was purchased, at their lowest possible offering prices, we found this highly touted class of IPOs yielded the exact same return on average, as the plain-vanilla indexes, like the Dow Jones Industrial Average. 

Innovative investors are adapting to the confusion. Almost universally, we found efficient value seekers moved away from analyzing vast swaths of “Big Data.” Instead, investors allowed themselves to be led by targeted human research and ground proofing to study reasonably-sized proxies for full-scale market data. Simpler indexes like the S&P 500, Nikkei 225, or FTSE 100 proved as effective as working through vast geographies of globally traded equities. 

Big Data was out. Artisanal information was in. 

Photo: Anete Lusina via Pexels.

Within these limited analytic universes, even the most algorithmic and quantitative investors migrated quietly towards analytic approaches that were more open to human intuition and narrative dimension. Strict government-required accounting measures, like operating cash flow and net profit, were no longer the benchmarks of value in 2020. Instead more qualitative measures, like free cash flow or economic profits, sought to bake in exogenic factors like the capital expenditures and opportunity costs. 

The relationships hidden in data were still being carefully calibrated. But those calibrations were given fresh room for broader considerations. 

But the biggest changes in how value was considered came in the prediction game. Where once time-tested, usually quarterly historical, data feeds were the points of origin for what the future might bring, last year saw the move to real-time data. For several months, during the early days of the Covid lockdown, traditional data from the likes of the Bureau of Labor Statistics and The Federal Reserve, were either incomplete or flatly unavailable. In their places, the valuation community began to incorporate ersatz data sources like the UKG Workforce reports, or online real estate figures from operations like RedFin.

A bullish stock market then pushed these new valuation methodologies to dig deeper for forwarded-looking productivity: “Who did more with what?” seemed to be the major question for the past year. Inventive stakeholders in the valuation community found intriguing ways to sniff for the efficient from the also-rans. Basic models of productivity, like revenue-per-hour of labor, or return on invested capital, were part of the valuation mix. But we also saw a new family of data inputs: Power usage, truck rolls, or the amounts of online computational storage were some of the more subtle predictive inputs that found favor over traditional predictors, like the price of crude oil or the unit volume of car sales. 

It seemed that investors who took the time to listen for the real story a company was trying to tell, were better prepared to understand what that company might be worth. 

Photo: George Becker via Pexels

But here was the trick: What mattered most was not absolute performance. True insight, at least for now, came from carefully calculated relative value. Truly astonishing bits of mathematics were used to quantify the rates of change in the relative values created by real time analytics, more qualitative data classifications, and human powered research. For the record – and the math-obsessed – the most clever models we saw, relied on nonlinear systems, including various families of differential mathematics and convolutional operations. 

We found that the truly effective investors – the ones did not overpay and had the discipline of when to sell – found ways to sensibly compare these newly engineered valuation tools to existing benchmarks found in wider markets. Those who truly thrived, invested in automating just enough of this educated intuition to keep their valuation perspectives in tune with often dramatic changes in the markets. 

As far as finding value in 2021? Our hunch is, the winners will be those who use enough intelligent machinery, mixed with qualitative common sense, to ask if they are asking the right questions at all.