Supply chain failures were treated as a novelty when they arrived. The mechanisms involved are well understood in the field and were largely predicted.
The bullwhip effect
The central concept, described decades ago and demonstrated repeatedly.
A small change in end demand produces progressively larger swings further up the chain.
The mechanism is that each stage orders to meet expected demand plus a safety buffer, and each stage sees only its immediate customer's orders rather than actual end demand.
So a modest increase at retail becomes a larger increase in distributor orders, larger still at the manufacturer, and enormous at the component supplier.
The same amplification runs in reverse when demand falls, producing the cancellation cascades that followed the shortages.
The semiconductor case
The clearest illustration and the most consequential.
Vehicle manufacturers cut chip orders when demand collapsed early in the pandemic.
Chip fabricators reallocated that capacity to consumer electronics, where demand had risen.
When vehicle demand recovered faster than expected, the capacity was committed elsewhere, and lead times for new capacity are measured in years.
Which produced a shortage of comparatively simple chips that halted production of vehicles costing thousands of times more than the missing component.
Why lean systems amplified it
Decades of inventory reduction had removed the buffers that would have absorbed the shock.
Just-in-time systems are genuinely efficient and depend on reliable supply.
Which is a rational trade when disruptions are rare, and it concentrates the cost into the rare events.
Firms had been optimising for cost with disruption treated as a low-probability event, and the probability turned out to be higher than assumed.
The container problem
A physical logistics failure independent of manufacturing.
Containers accumulated where goods were consumed rather than returning to where goods were produced, because trade flows became unbalanced.
Port congestion meant ships waiting to unload, which removed capacity from the system, which worsened congestion.
Freight rates rose by an order of magnitude, then collapsed when demand normalised and new ships ordered during the peak arrived.
That overshoot is the classic pattern in shipping, which has cycled this way for a century.
What firms have actually done since
Increased inventory, reversing decades of reduction, at a real cost in working capital.
Dual sourcing, qualifying second suppliers for critical inputs, which costs more and takes time.
Mapping supply chains further upstream, since many firms discovered they did not know who supplied their suppliers.
And nearshoring in some cases, though the volume of actual relocation has been considerably smaller than the volume of announcements about it.
The resilience trade-off
Every resilience measure costs money in normal conditions.
Which means the pressure to reverse them builds as disruption recedes, and the discipline required to maintain them is largely a governance question rather than an operational one.
Historical patterns after previous disruptions suggest buffers erode within a few years.
What to expect
Concentration risk remains in several categories — specific components, specific chokepoints, specific regions — and is well documented.
Geographic concentration of critical inputs is now a stated policy concern in most major economies, with subsidy programmes attempting to diversify production.
Those programmes take years and are unlikely to remove the concentration entirely, since the concentration exists for cost and expertise reasons that subsidies only partly offset.
Chokepoints
Specific physical locations where an enormous share of trade passes and where disruption has outsized effects.
A small number of canals, straits and ports carry a disproportionate share of global maritime trade.
Which means a single blockage, closure or security problem at one of them affects global costs within days.
Recent years have demonstrated this repeatedly — a grounded vessel, drought reducing canal transits, and security disruptions each produced measurable global effects.
Alternative routes exist and add substantial time and cost, which is the definition of a chokepoint.
Concentration in specific inputs
Beyond geography, several critical materials and components are produced by very few firms or countries.
Processing capacity for several critical minerals is heavily concentrated regardless of where the minerals are mined.
Which means diversifying extraction does not diversify supply if processing remains concentrated, a distinction frequently missed in policy discussion.
What buyers can actually do
Mapping beyond the first tier, since most firms know their direct suppliers and not the suppliers behind them.
Identifying single points of failure explicitly rather than assuming diversity because several suppliers exist.
And holding inventory of the specific items whose absence stops production, rather than uniformly across everything.
Contracts and force majeure
Where the disruption became a legal question.
Supply contracts generally contain clauses excusing performance in defined extraordinary circumstances, and whether a pandemic or a port closure qualified depended on precise wording.
Enormous volumes of dispute followed, and contract drafting since has become considerably more specific about what is covered.
Which is a lasting change and one of the few that is unlikely to erode.