Howard Marks on Crisis Response

I have used the title Nobody Knows three times now — first in October 2008 as the financial crisis was accelerating, again in March 2020 as the pandemic shut down the global economy, and now in 2025 as the new tariff regime disrupts assumptions that had been built into global supply chains and asset prices. The recurrence of the title is not laziness; it is a reminder that the most important macro questions are unanswerable in real time, and that the right response to uncertainty is humility rather than forecast. What I observed in April 2025 was that the announcement of sweeping tariffs represented a fundamental reordering of the global trading system that has been in place, in one form or another, since the end of the Second World War. Whether the policy will be sustained, modified, or reversed is unknowable. What is knowable is that the assumptions embedded in many asset prices — supply chains that depend on free movement of goods, cost structures that assume low tariffs, and growth models that assume continued globalization — were suddenly subject to material revision. Risk management in such an environment begins with the admission that the range of outcomes has widened. When the range of outcomes widens, the right response is not to make a more confident forecast but to demand a larger margin of safety. The investor who is uncertain about the path should pay less, not more, for the assets that depend on a particular path being taken.

When the financial crisis began in earnest in the late summer of 2008, the question I was asked most often was whether the situation would stabilize or get worse. My answer then, and the one I have continued to give in every subsequent crisis, is that I do not know. Nobody does. The honest investor admits this rather than dressing up uncertainty in the language of conviction. What I could see was that the structures which had been built on the assumption of permanent liquidity and ever-looser credit were beginning to fail. The unfreezing of credit markets depended on confidence, and confidence is the asset that disappears fastest when it is needed most. The mistakes that brought the system to the brink were leverage, complexity, and complacency about correlation; the solutions would have to address all three. Risk management in such an environment is not about forecasting the bottom. It is about avoiding irreversible loss. The investor who survives a crisis intact has the optionality to participate in the eventual recovery; the investor who is forced to liquidate at the wrong time does not. The preservation of capital through the worst of the panic is, in retrospect, the precondition for the returns that came after.

The crisis revealed how thin the layer of true liquidity actually was. Instruments that had traded daily in normal markets became untradeable. The bid-ask spread that had been a rounding error became a chasm. Capital that had been committed on the assumption of roll-over financing had to be redeployed at any price the market would bear. The lesson is that liquidity is a regime-dependent asset, and the regime that produces abundant liquidity is not the regime in which you need it most. What we did at Oaktree during that period was deploy capital into the dislocations. The opportunity set was the widest I had seen in my career — distressed debt trading at prices that implied default rates several times any reasonable estimate, structured credit that had been marked down mechanically, and senior secured loans trading at deep discounts to par. None of these would have been available at those prices in any other market environment. The decision to buy aggressively required capital, conviction, and a tolerance for being wrong in the short run. We had raised a meaningful distressed debt fund in 2007 and 2008 that gave us the dry powder to act. Without that capital pre-arranged, we would have been unable to participate. The lesson of 2008, as of every prior crisis, is that the time to raise capital for distress is before the distress arrives.