Known Unknowns


Welcome to Known Unknowns, a newsletter that—just like everyone else—is making bold predictions based on incomplete data.

Risk and uncertainty

We are living in uncertain times, but this is different than living in risky times. I had an op-ed in the Wall Street Journal last week about the difference between risk and uncertainty. The future is always unknown, but risk is a measurement of all of the myriad things that can happen. And that measurement comes from data, because you are facing a situation somewhat similar to something that has happened before. The good thing about risk is that it can be managed. You can weigh trade-offs and decide whether the risk reduction strategy is worth the cost.

The problem with the coronavirus is that we are in the realm of uncertainty. Uncertainty refers to the risks that we can’t measure because they are so unexpected and unprecedented. We still know so little. It seems that the virus lives on cardboard for days, but can you get it from touching an infected box? How many asymptomatic carriers are out there? What’s the death rate after you control for age, pollutions levels, history of smoking, etc.? Is going for a walk an acceptable risk?

None of these things are truly knowable. Any projections that we see of curves, time lines, or predictions as to how this will end are all based on incomplete data and heroic assumptions (but to be fair, they had no choice), so we shouldn’t be surprised or overly critical if they don’t actually pan out.

Only one thing can turn uncertainty into risk, and that’s good data. Thus we need lots of it, and we’ll get it, over time and after lots and lots of testing. Data can give us reliable estimates, and then we can weigh the trade-offs of opening the economy—or even walking your dog outside.

As we see more cases reported, it is truly terrifying to see the scope of this virus, but I also take some comfort because it also reflects that more testing is being done—and that means that more certainty is to follow.

CARE is not a stimulus

I remain a fiscal stimulus skeptic. The evidence that government spending can meaningfully increase demand, given the cost, is mixed. That may be because it’s hard to choose projects that increase growth. But, like many skeptics, I think that this is the time for the government to spend big. However, it’s not about boosting demand; it is, in my opinion, time for the government to act in its role as the insurer of last resort.

The spread of the coronavirus was a very large, uninsurable shock. Self-insurance for a virus shutting down the economy for months would have been very expensive and inefficient. It’s easy now to say that every business should have six months of emergency funding in cash, but if they actually were able to do that—and did—we’d complain in good times that they don’t invest enough.

The government is in a better position to provide insurance, as they can diversify across individuals and different cohorts (by borrowing now). That’s why we should not call the CARE act a stimulus. Because it’s not—it’s insurance. And this isn’t just about terminology. If we see this government policy as insurance, it can help us to craft the best and most efficient policy. There is no point in stimulating demand now, even if you think it might work. But we do want to keep firms in business and help people continue to pay their bills.

I agree that now is not the time to be timid, and interest rates are negative, so borrowing is cheap. But that doesn’t mean we shouldn’t be smart about spending. Rates are low now, but if the aftermath of the virus is less global integration, then there will be fewer buyers of American bonds, which means that rates could go up in the foreseeable future.

I had an interesting conversation with Ricardo Reis about who should pay for any bailouts. Should we just give firms money that they don’t have to pay back, or should we offer subsidized loans? If there are loans, then they pay for their bailout, but if they are transfers, then future generations do. You can justify wartime bailouts because future generations enjoy the peace dividend. But whether or not that’s true right now isn’t as clear.

Corporate and muni bonds

The Fed’s actions are also best understood as insurance. They are not trying to juice the economy—rather, they are trying to keep it on life support. And that’s why they are buying corporate and muni bonds. I never thought I’d believe that it was a good idea, but if those markets freeze—and there’s a good chance that they could—then there could be a need (or a political call) for explicit bailouts. Keeping those markets as liquid as possible is a much better idea. I’m especially worried about the BBB corporates who face the risk of downgrade and then a massive sell-off. Yes, I know that markets have already priced that risk in, but that assumes it’s a normal measurable risk.

I’m not sure if pausing ratings is the right idea, though. After all, we want more risk and less uncertainty. True, many pension funds and insurers can only own investment-grade corporates, and a downgrade forces them to sell. But it seems like those requirements could be relaxed for now, and that would be better than making this market totally opaque.

Also, things could only get worse for muni bonds. One day, firms will eventually get back to business. But states and municipalities face a more uncertain future, with big bills from fighting the virus and less tax revenue. I don’t even want to think about their pension funding ratios.

Also on risk: I suggest reading Ricardo’s thoughts on the Fed extending dollar liquidity swap lines to emerging markets. Keeping countries rich in dollars is critical right now, especially for countries that have so much dollar debt. It also helps to ensure that the dollar will continue to be reserve currency—a privilege that we’ll miss when it’s gone. And keeping these countries afloat will be critical to keeping the virus at bay in the coming months, but it also exposes the Fed to risks that push the limits of its power. There are no easy answers there.

The future

No one knows how the world will change. I tend to believe that humans are adaptable. Many people in the world have lived through forced shut-ins for much more extreme, scary reasons, and have returned to normal one day. I think people will appreciate in-person work and lectures more than ever before when this is all over.

But I do wonder if the pace of globalization will change. Like many economists, I see many gains from more global integration in the form of more diversification and the comparative advantages of trade. It’s how so many people escaped poverty and we got many new consumer goods very cheap. The iPhone really is a miracle. I know, I know—productivity and stagnant growth—but new innovations sometimes have long lags in productivity. You can’t deny how much quality of life has improved.

But we are now seeing that globalization also means more tail risk, both in terms of virus spread and supply chain disruptions. It’s like buying an annuity that removes market volatility in exchange for taking on tail risk. We often ignore tail risk until it manifests. So, could we expect more nationalism and less trade? That would mean less growth, higher prices, and maybe higher interest rates.

Perhaps for a while there will be less trade, but people often become complacent about tail risks over time, even if they have experienced their downsides.

In other news

Australia’s boom is finally coming to an end

Until next time Pension Geeks, stay well!