Why alternative data is here to stay

Global macro, which can be adequately defined as the analysis and prediction of economic, political and financial market developments, has also seen some early movers adopt alt-data.

However, the uptake has been slower, with legacy players dominating the two main variants, discretionary and systematic.

COVID-19 has changed that. Through late Q1 and early Q2 demand for alt-data stepped up dramatically. Demand has trended higher since, and at a much more accelerated rate than before COVID-19. This should not come as a surprise.

Global macro, at its core, is about making investment decisions. If those decisions are being made with an inferior information set, an asymmetry will develop over time, and ultimately degrade investment performance and the management of risk.

There will be exceptions to the rule. But not many. Even long-term asset allocation frameworks were tested through the early onset of COVID-19, in terms of maximum drawdown metrics.

The same can be said for risk parity, where measures of equity beta fell sharply through March, and are yet to recover substantially.

An alternative way of framing this is in terms of the classic distinction between risk and uncertainty. Those who embraced alt-data early on during COVID-19, were able to manage through the volatility of March in the context of quantifiable risk.

The moves were extreme, but alt-data, both in terms of COVID-19 and measures of social mobility, helped to fill the void and empower decision making. Those without such resources were facing Knightian uncertainty.

A story in two parts

At Exante Data, we saw two distinct phases. The first order challenge was the collation of cases, testing and tracing data, along with the early specification of hospitalisation constraints.

Providing a layer of human interpretation was also key, particularly early on, when the comprehension of basic epidemiological modelling was poor.

Related to this, in early February we started to compile real-time mobility metrics, principally from Baidu, that tracked China’s return to work.

These data were used by the Imperial College of London’s team in publishing one of the earliest studies that confirmed it would be possible to lift social distancing restrictions without necessarily driving a resurgence in the virus.

The demand for these sorts of insights has continued throughout, and we have found some surprising results along the way. For example, restaurant patronage in the US has been the most reliable proxy of social distancing.

In contrast, the Black Lives Matters protests did not move the needle. As a public service, we publish an assortment of these data each day on our Twitter feed.

The second main focus has been social mobility, defined broadly, as it pertains to the economy and financial markets themselves.

Our most aggregated measure of global lockdown, traffic congestion across 411 cities in 56 countries, bottomed out on April 6, and rebounded strongly through May and June, before plateauing.

While global equity markets based with central bank interventions in late March, this sort of measure has provided strong real-time corroboration of the rally back in equity, credit and commodity markets.

At a more granular level, we have shown alt-data to lead PMIs and retail sales prints. The bounce in oil prices, against a backdrop of high global inventories, also makes much more sense in the context of miles being driven.

Where we have seen less explanatory power is in the context of foreign exchange markets, where our capital flow tracking is still much more relevant than either COVID-19 or social mobility.

From our clients perspective, the main attraction of alt-data has been its timeliness. However, we have also emphasised the importance of the higher frequency and the broader sample, as compared to traditional economic data.

Will data remain public?

Where this story gets sticky is in terms of institutionalising the advances that have been made.

Many of the most important data sources that are now available are from tech giants, and these may revert to the private domain if and when COVID-19 runs its course. Baidu, for example, withdrew its data sets in early May.

There are also limitations that arise from privacy regimes, that vary considerably by jurisdiction. Europe’s General Data Protection Regulation marks one extreme, with China’s multi-faceted surveillance apparatus at the other.

Data sovereignty is also an important issue, which goes to where alt-data resides and the forms in which it may be published. Individual licensing is a further commercial consideration.

Beyond this there is the question of institutional rigidity.

While investment funds and corporates have sought out alt-data, the embrace has been more grudging within central banks and public policy circles more generally.

There has certainly been some engagement, and some lip service paid as well, in terms of formal monetary policy decisions.

However it is unclear to us whether governments, broadly defined, have the interdisciplinary capability to gather, analyse and nimbly bring alt-data to bear for the most timely and relevant problems faced by policymakers.

To be sure, there is a longer time horizon at play in government. However there is also a deficit of technical skills, and surplus of bureaucratic constraints.

Where there has been a clear about-face is by the big sell-side banks. These are the same banks that confidently dismissed COVID-19 as a SARS-lite episode in February, in projecting a V-shape recovery into early Q2.

The sell-side is not in the habit of making apologies. But this was a bad miss.

They have filled the breach by embracing alt-data in toto, with many banks now publishing daily monitors.

This, in turn, will have important self-enforcing implications, in terms of the availability of alt-data going forward, and its pre-eminent role in global financial markets.