Sometime between March and April, the economy will be in recession or depression. Guaranteed.
The real strength of these predictions is not the fact that such things come-forecasted by almost every economist for ages, nor a supposed ballpark of the month where some flexibility will save your ass, as if there were some special uncertainty about the month, but never related to the economy which gets none.
Nor is it the amount of luck such extraordinary collapse will take. Disrupted energy, transport expenses, fertilizer shortages causing agricultural damage, consumers just exhausted, an artificial-intelligence bust... All of that, and maybe more, on top of already strained public finances.
What I address here is not the extraordinary luck a collapse case will require. Neither am I talking about the lifespan assigned to the word This. You might be surprised to know there’s such a difference between announcing an oil shortage, falling stock prices, or household financial distress as causing tomorrow’s predicted recession and today’s depression.
When we talk about avoiding a decline in activity, it’s important to note that this doesn’t mean experiencing some type of brief, vicious cycle of prolonged collapse involving employment, production, and a financial crisis. At its worst, it will manifest as obligations that cannot be met and losses spreading. Perhaps a sequence of expensive groceries feeling miserable. But this is not the case we see now.
The situation will be more difficult to see when we are moving in and out of conditions like the ones portended by the above description. It is much clearer when you can put up a deadline in spring when some of these things happening to squeeze production, consumption, or financing. What could add up to damaging effects. But we must also stress that some of this has to give. In trying to avoid households buying less, businesses accepting lower margins, or investors all taking losses, it will be down to governments, making expensive interventions that leads to a somewhat impossible situation. Our forecasts will typically present us with some details on this front.
Start with the word blocked
Conversely, if energy-related effects ever could make conditions like major shipping routes remarkably unavailable or simply moving into a less productive state, surely beliefs that alternative pipelines or similar projects could lead to closing down petrol stations could suddenly turn more reasonable. Fuel not only is critical to freight, but sustains agricultural operations and industrial production too. Stalling businesses from taking care of their efforts would force increasingly difficult choices as inventories that are accessible gradually would run down.
Consider what blocked means. It may not yet be clear what the word means in relation to duration, but it almost certainly means greater than brief, therefore. Does it mean all, or could some supply be moved, albeit at greater cost? Who are your buyers assuming you rely on the word for your forecasts and have kept hidden these conditions that are central to your assumption?
What are you assuming? Could some supply be redirected? what if particular users substitute another input, what if production continues albeit at higher prices? All of these conditions would suggest that your forecasts could be more accurate and consequently ultimately much more valuable. Assume you are just a little slow in preparing them. Expensive however unsuitable you may assume the level of relief that is being searched for.
Popular as the quote became, an alarming rental for a particular class of tanker, on a particular route under a particular contract for a particular date, does not preclude it being the price or rental offered for that particular journey. It will not however, necessarily match the charge for transporting every cargo or predicting the eventual increase it will cause in the supermarket aisle. You can be certain however that the household receiving the higher grocery bill will not care, you had better be certain that anyone forecasting that bill is.
The signals and noise problem is particularly evident in this cycle, for two reasons. The price spike has hit energy and transport (both major inputs to fertilizer) so it’s possible that the connected description of the same shock could lead to spikes on all three fronts. Counting the shocks separately could actually amplify them, and independently confirm what is likely an inevitable cascade.
This is a commonplace in the dynamical systems we’ve been looking at for the last few days. Simply ordering and timing the shocks allows us to calculate their effects on the system, and determine the properties of the system that will be threatened by a collapse.
We can’t do that in our situation, because we don’t have a deadline. And while it is useful to think of plausible sequences, and expect that some of them will actually happen, it doesn’t prove anything. Lastly, it is dangerous to make the same expectations of dynamical systems of our thinking on sequences, instead of on the ordering of events.
And the household still has bills
The impact on families and households is obvious and it is yet to be seen how much it will constrain consumer spending. After a certain level the aggregate amount that can be borrowed by consumers will be limited, and lending conditions are becoming more difficult. Besides, something will have to move in the household budget when you are already close to the edge. This is true, but consumers as a whole is not a single household. The real question is how many people are having an immediate problem making their payments, how many are feeling the pain of higher essential costs. Of those, how many will cut back on discretionary purchases and which creditors will be affected. Private hardship without a doubt contributes to a wider contraction when there are sufficient numbers, but we have yet to see how many will be financially exhausted.
It’s better to run a household than an economy: most households should be able to get by, and not every household that is forced to scrimp and survive will mean that the economy will no longer be producing and transacting. It’s true that there will be particular households where the pressures are intolerable, and one could make a forecast of severe hardship in the future.
But most of the data we look at to figure out what is going on in the economy is backwards-looking. So even though it is impossible to predict the precise timing of any possible collapse, those of us who would like an economic prediction need a better idea of what to look for.
How do we know we are in one in process?
The most important point is that economic activity has not ceased completely. Families might be postponing new purchases or waiting to make investments in their business, but as long as they continue to buy essentials there is some economic activity, even if the displacement of other spending means considerable damage is being done.
At least for survival, then, we have defined our standard.
It would be silly to hold the economy to such a low standard for prosperity in 2028: we’d be better off looking for firms that have failed and investments that have disappeared.
But these will still be ones that will cause households to suffer lasting losses and create serious political problems.
Who owes whom?
Much of the discussion of the ill effects that a proposed artificial-intelligence bust would have focus on comparisons to earlier bubbles. There are certainly lessons to be learned from those. But the practical implications of what might happen next depend in part on which aspect of the business cycle is guiding comparisons. Take equity valuations, for instance. In a traditional model, large gains from investment commitments translate into operating losses over time. But in this case the pressure on the financial system is very different.
A traditional business cycle holds that any industry causing so much damage must have a bunch of participants that are all relying on the same source of funding. We don’t have that component in this cycle. For one thing, much of the equity valuations are simply priced into expensive shares. Speculative borrowing is also absent. So rather than worrying about obligations owed to some sort of financing (who owes whom?), investors ought to be much more focused on the question of what sorts of projects they’re committing funds to, and how much money they stand to lose.
A fall in asset prices could trigger a shortage of cash needed to meet obligations only if it can be shown that one or more obligations will not be met. Otherwise, any disappointing investment could be promoted to the status of a financial emergency. It matters if this turns out to be an investment boom that merely ends badly (because of a fall in confidence, perhaps) as opposed to a boom that goes poorly for other reasons. A really rich appraisal of past experience, perhaps with some better resemblance than mere mood, would be helpful in this respect.
The buyer steps back. Then what?
A rising tide of public debt could one day drown an economy, crushing it under the weight of servicing ever-growing debt. It’s a rare fact, though, that draws attention away from the public in pain, as it shrouds in darkness the problems that add to its burden. What happens when the demand for foreign buyers goes down? How high will financing costs have to get before demand goes back up? Does it really matter when the foreign buyer has so many choices of securities? What other buyers will be attracted and at what price? What affect will changes in financing costs have on spending, taxation, and credit conditions?
When one foreign buyer steps back, it’s hardly a door closing on the forecast.
But simply articulating all potential adverse scenarios does not justify assigning certainty that they will all materialize. Perhaps even a severe energy shock will not materially relax household spending, relieve speculative investment, or bring public finances under additional pressure. Yet the potential exposures could also react with one another and increase the scale of themselves, on the theory of compounding.
Assigning certainty to every single aspect of this issue, whether we are identifying size of exposures, insufficiency of buffers on hand, futility of all potential responses, requires significantly more evidence than merely listing a long inventory of hazards. We cannot simply assume that an inventory is complete, let alone have certainty that the evidence will point in a certain direction.
Suppose spring comes and goes
What, then, allows for continued economic activity, even into 2028, in the absence of any adjustment? Simply the fact that an adjustment will mean people and businesses plan on less consumption, some businesses will close or shrink, investors take financial losses, and governments pay the costs, not a comfortable adjustment, but how we continue when some parts of the economy are doing badly.
An inevitable-collapse forecast needs to explain why that response is insufficient; merely describing the pressure leaves that task unfinished.
No one could’ve doubted that it would really be a miracle if we were to emerge from another year without suffering even one serious economic loss, if not a recession. Nor is it essential that every arrangement now in force should survive. Many arrangements and enterprises are bound to fail; businesses, projects, and households will cancel their plans, suffer losses, and, ipso facto, in all probability, never recover. This may be serious, but the significance of these failures can easily be assessed by the way we consider or characterize their possible outcomes.
Their significance becomes serious, if we start qualifying the expectation of collapse as final. As long as the expectation is simply an observation, the logical structure of the argument is correct. It is as yet undecided whether in future there will indeed be confirmation of the forecast of collapse, or whether at the last minute it might be called off. And on what basis could we decide this? If the conditions deteriorate, that is confirmation; and if they improve, it will be temporary, so the deadline can simply be moved.
If spring arrives and finds people still able to pay for necessities, businesses not yet cutting jobs, and families going about their lives without purchasing the things they need, it will not mean we avoided the promised collapse. It means we have bought more time. But that answer cannot explain every possible outcome forever. If things getting worse confirms the forecast, and things getting better merely postpones it, what would ever count against it?
There may still be families struggling with bills when the deadline passes. Their hardship deserves attention whether or not the promised collapse arrives. It would be a miserable result if an overconfident prediction gave everyone else an excuse to stop listening.
References:
U.S. Energy Information Administration, World Oil Transit Chokepoints