Handling Covid-19: an extreme case study in managing the growth vs. risk dilemma
We’re currently living out an unprecedented crisis caused by the outbreak of the Covid-19 virus, which impacted directly or indirectly billions of people around the world. Multiple governments have imposed strong measures such as travel bans and shelter-in-place orders to reduce the spread of the virus.
We fully support the measures taken by the authorities and understand the weight on their shoulders: to overcome this very challenging situation with high uncertainty. Here at Tint, we spend our waking hours helping companies manage their insurance programs, balancing and optimizing their growth and risk trade-off. As this worldwide pandemic has unfolded, we realized that governments around the world are doing the same thing, albeit on a much larger scale. Ultimately, they need to use whatever information is available to segment the risk as much as possible so they can impact the lowest part of the population while reducing the maximum potential loss.
The growth vs. risk framework
A couple of months ago, we published an article introducing the risk vs. growth framework to maximize the profitability of a company. There is a clear trade-off between, on the one hand, accelerating growth by reducing customer friction, and on the other, controlling risk by increasing friction. For example, a platform can stop verifying its users to increase the number of transactions that are completed. It’s likely that sales will go up, but so will the number of bad experiences and losses related to bad actors acting freely. In this case, the additional volume brought by the lax verification policy may not be good for the business.
Now, consider the opposite example: the platform decides to manually verify every single transaction and uses a policy that is so strict that it stops all transactions. It succeeds in eliminating all the risk, but it has also halted all revenue, which is also unsustainable.
The answer to this dilemma is to find ways to segment the transactions based on a variety of data points so the company can identify high-risk transactions without impacting the majority of its transactions, which tend to be low risk. In our experience, losses are normally concentrated in a small minority of high-risk cases.
Applying the framework to the Covid-19 pandemic
Governments face the same trade-off: minimize the short-term impact on businesses and people to keep the economy working, the growth part, while reducing the risk of catastrophic losses of lives in the future due to the collapse of the healthcare system.
There are two aspects of the Covid-19 virus that aggravate the issue: a) the spread of the virus is exponential, so the authorities don’t have a lot of time to act; b) the virus may be symptomless, so it’s hard to identify the carriers and isolate the minority of high-risk cases from the majority of low-risk ones. There is not enough data that could be used to build rules or models that could segment individuals and allow the friction to be directed only to a small part of the population.
Governments have tried to use location as a proxy to segment users at the beginning of the pandemic. Countries have issued travel bans from/to places that were affected early, like China, Italy, and Iran. This is an example of a risk segmentation using a simple rule-based on one dimension (geography).
But as it often happens with companies dealing with complex risk threats, a simple rule is not enough to control the spread of the virus. This geography-based rule may have slowed down the contagion for a bit but it didn’t stop it. The countries that didn’t have massive scale testing and traceability (activating surveillance systems), such as the US and most European countries, decided to give up on this segmentation and raised friction by ordering that most non-essential services and businesses shut down for weeks, which helps in the fight against the virus but has material economic costs. Governments are stepping in to inject trillions of dollars in their respective countries to try to alleviate this economic impact.
In our framework, the US and Europe are equivalent to companies creating a lot of friction for all their users, which has a very large impact on the revenue. It manages risk at the expense of growth. This is exactly the right decision when authorities can’t isolate the risks and the stakes are too high. The actions deployed are the only way to control the pandemic and to avoid a much higher loss in the future.
On the other hand, South Korea, Hong Kong, and Singapore were able to control the Covid-19 outbreak without strict lockdown strategies. How did they do this? In two words: risk segmentation. They tested widely for the virus and leveraged data from mobile phones, cameras, and credit cards to identify and quarantine the suspected cases quickly. This segmentation allowed them to apply friction only to the small minority that was affected by the disease.
Here is how the measures adopted by different governments fit into the growth vs. risk framework:
Therefore, data-driven ways of segmenting risks are critical to reducing the economic and physical hardships that most people affected directly or indirectly by the virus go through. We are hopeful that as governments, companies, and healthcare systems react to and learn from the pandemic, they will be able to utilize technology to, slowly, but surely, effectively manage the risk and get the economy going again.
Thank goodness companies have it easier
In comparison with the global pandemic, the risk problems that companies face are much smaller so it is possible to use data and models that segment their risks and let them grow while reducing losses. Some businesses managed to create effective tech-enabled risk management and insurance strategies that built trust. The advent and growth of the Sharing Economy is an example, as strangers engage in mutually-beneficial relationships on a scale that was not possible without technology. This is also evident in the insurance industry, where modern products that rely on data and real-time segmentation are being launched every day to protect customers.
We are likely to live for weeks, if not months (years?) in a world where uncertainty is high. Companies will be pressured to cut costs and find profitable growth. For startups, the VC-fuelled growth-at-all-costs philosophy is over. Therefore, the choice between growth and risk will be replaced by the need for profitable growth.
As a company that uses AI-powered technologies to help companies reduce losses and monetize their risks, we feel honored to be in a position to contribute. We know it is cliche, but the best companies are forged during crises as they figure out how to optimize each aspect of their business model.
