Supply Chain Resilience

The recommended pause in administering the Johnson & Johnson vaccine will have modest near-term network-level impacts. For a host of reasons, J&J has not been expected to begin releasing large numbers of doses until late April and May. This week (April 12-17) the J&J allocation for states was sized — pre-pause — at about 700,000 doses compared with almost twenty million doses for the combined Moderna and Pfizer vaccine flows. Below is how Bloomberg has estimated production outputs. J&J is continuing vaccine production. If and when J&J is un-paused total flows will catch up.

On March 31 the seven-day moving average for new hospitalizations in the United States was 4969. On March 16 the seven-day moving average was 4701. This is the wrong direction. Will this increase persist? This is the sort of slight upward curve that the Europeans saw starting in late February.

A journalist has asked several to write — in 300 words or less — what the pandemic has taught us about supply chains. Here is how I responded:

What did the pandemic teach you about supply chains?

The pandemic has (so far) mostly emphasized essential characteristics that tend to be taken for granted in less challenging contexts.

The pandemic taught me that organizing around demand is fundamental. Where and when manufacturers, carriers, customers, and even competitors collaborate to calibrate around demand, flows can adjust to new needs much more quickly.  Where demand is neglected or miscalculated or just not known, supply misses the mark more and more, troubles accumulate, velocity is reduced, so volumes are not maximized to fulfill demand.

The pandemic taught me that where the scope and scale of demand can be confirmed (and effectual demand continued over time), extraordinary creativity is possible in terms of streamlining and maximizing existing production capacity (much more than I expected). But I also was reminded that production capacity is constrained by access to capital, stubborn issues of time, and — especially –perceptions of sustained future demand.  

The pandemic taught me that robust, diverse, agile freight systems can multiply and divide every other supply chain capacity. When freight assets — surface or maritime or air in whatever mode — are most flexible, then the whole supply chain is more resilient. Persistent freight problems will undo every other advantage and double any other disadvantage. When and where freight channels are flowing, problems with production or demand or whatever can usually be mitigated.

The pandemic taught me that humans — even smart, self-aware, and alert humans — tend to discount the likelihood of any major shift in prior experience, even when the evidence is obvious and racing toward us.  Probably related: the pandemic taught me that smart, self-aware, and alert humans tend to focus on well-defined problems or opportunities and tend to discount ill-defined issues, no matter how potentially consequential. (Yikes!)

The current situation in Europe, India, and Brazil each (and all) demonstrate the continuing risk of another surge in disease.

US hospital admissions related to covid continue to ease. This indicator is our most rigorous signal of covid’s “demand-pull” on the healthcare system. In my judgment fatalities, while even more rigorous, signal demand cessation or, perhaps, demand intensity.

US case counts are suggestive and worth watching, but substantially undercount actual “consumption” of covid.

Especially in comparison with much of Europe, India, and Brazil, these US data indicators are clearly more positive. But current case counts (at least in the US with its anemic testing/tracing system) typically lag viral flows by a matter of several days. New hospital admissions typically reflect disease conditions two or three weeks prior. In either case, once case counts and hospitalizations are surging it is too late to stop the surge. Post-hoc we can only contain.

Searching for an ex ante indicator, my best — if very rough — tool has been cell phone mobility data. It has seemed to me (but much more diligent data analysis is needed) that when and where human circulation has remained at three-quarters pre-pandemic levels, covid has been much more effectively demand-managed (two-thirds is even better, the ROI beyond 2/3s has, so far, seemed negligible).

So… from a supply chain perspective, US behavior over the last three weeks is cause for concern. Can US vaccination rates mitigate the risk resulting from this increased circulation? I am much more confident that vaccinations combined with ten percent less human circulation (or even less) than shown below would avoid another demand-surge on the healthcare system in mid-to-late April.

A recent White House Executive Order notes, “The United States needs resilient, diverse, and secure supply chains to ensure our economic prosperity and national security.”

Supply chain risks can be reduced or transferred, but many risks cannot be avoided. For demand and supply networks to fulfill their purposes, a wide range of insecurities are innate to purpose and function.  Supply chains cannot be fully secured and remain supply chains.

For some purposes it is possible to create highly secure manufacturing capacity, transportation modalities, and effective deployment of technologies and related services. But this process is unlikely to benefit from high velocity, cost-amortized-over-volume, self-actualizing, demand-fulfilling networks, which are among the key characteristics that differentiate “supply chains” from other less efficient forms of supplying demand.

In many ways, supply chains can be resilient and diverse or we can create secure means of supply. When we create secure processes, we should explicitly recognize related constraints. In some contexts, more security is entirely justified, but it will always travel with higher costs, less agility, and usually with reduced speed.  

We should also explicitly acknowledge implications of insecurity in demand and supply networks. Given the innate insecurity of supply chains, systemic resilience is an essential characteristic for the most human-critical and nation-critical supply chains. Supply Chain Risk Management, strategically conceived, is primarily the cultivation of Supply Chain Resilience. There will be failures.  Recurring, relatively small failures typically characterize the most resilient networks. As high volume, high velocity networks are optimized, the circuit-breaking characteristics of small failures can often be squeezed out. Over time an increased risk of cascading, large-scale failures can be — unintentionally — built-into demand and supply networks.

Supply Chain Risk Management is aware of this tendency, actively monitors flows for fitness and systemic risk, takes action to mitigate risk in advance, and prepares to effectively contain and recover from catastrophic cascades.

Five Core Principles of Supply Chain Resilience

Supply Chain Risk Management advances Supply Chain Resilience by cultivating five core principles:

Principle: Resilient demand and supply networks are diverse, involving many different places, people, processes, forms of competition, forms of collaboration, varied conveyances, widely distributed sources, multiple pathways for demand and supply.  Concentrations (hourglass structures) facilitate volume, velocity, and therefore robustness (see below) but over-concentration can amplify systemic vulnerability. Practice: Observation: identify and map sources, links, relationships, proportional flows, and current/recent deficiencies. Orientation: Ecosystem flow (more) and fitness (contrasted to any particular species). Decision: Is diversity being fostered or impeded?  Where is diversity most at risk?  Where is less diversity most consequential? Why?  Action: Enhance diversity.

Principle: Resilient demand and supply networks feature abundant feedback. Demand signaling is well-facilitated across the network. Major market conditions (e.g. input costs, supply availability,  freight market performance, transportation conditions, etc.) are known.  Major players know each other and often communicate together either directly or through many trusted intermediaries. Practice: Observation: identify and map feedback loops. Orientation: Ecosystem flow and fitness (contrasted to any particular species). Decision: Is feedback being fostered or impeded?  Where is feedback most anemic?  Why? Where is this anemia most consequential?  Action: Enhance feedback.  

Principle: Resilient demand and supply networks feature wide-spread (diverse and feedback informed) agile self-organization.  A wide-range of individual, largely independent players have easy access to the network, participate in feedback loops, and are able to explore — and exploit — network fluctuations on their own decision and at their own risk.  Practice: Observation: identify and map indicators of self-organization. Orientation: Ecosystem flow and fitness (contrasted to any particular species). Decision: Is self-organization being fostered or impeded?  Where is self-organization least prevalent?  Why? Where is lack of self-organization most consequential?  Action: Enhance self-organization.  

Principle: Resilient demand and supply networks feature agile and effective adaptation.  When many diverse players are constantly signaling each other regarding network shifts and each player has significant freedom to explore and exploit these shifts, then myriad individual choices will filter failures and successes and in this way shift the entire network toward more effective behaviors. Practice: Observation: identify and map adaptive behaviors. Orientation: Ecosystem flow and fitness (contrasted to any particular species). Decision: Is adaptation being fostered or impeded?  Where or when is adaptation less common?  Why? Where or when is lack-of-adaptation most consequential?  Action: Enhance adaptation.  

Principle: Resilient demand and supply networks are robust: Resilience can deeply abide in very small, isolated (secure?) contexts.  But the more connected the context, the more resilience depends on robust scope and scale. Large systems — that are also diverse, feedback abundant, self-organizing, and adapting — will resist fundamental (or catastrophic) change, even while constantly changing through small failures and accumulated innovation. Practice: Observation: identify and map scope and scale. Measure pressure-levels of flow across the network. Orientation: Ecosystem flow and fitness (contrasted to any particular species). Decision: Are scope and scale growing or diminishing?  Integrating or fragmenting?  Are flows being widely facilitated or are some channels and sections being shed?  Why? Where or when does robustness seem to compete with resilience?  Where or when does robustness reinforce resilience?  Action: Enhance resilience.   

Operationalization of Supply Chain Resilience (or Practice Supply Chain Risk Management at the Ecosystem Level)

The matrix suggested by the preceding is so complex and dynamic that it is not currently possible to identify and map anything close to full flows.  Even meaningful snapshots are difficult and treacherous. Given current conceptual and technological capabilities it is helpful to set meaningful boundaries  for engagement. One of the most important outcomes of the six assessments set out in the Executive Order could be specific targets for operational engagement.  While there are clearly better or worse targets, the issue is not so much finding the most consequential target(s) as choosing a plausible set of targets with which to begin data-gathering, relationship-building, analysis, and synthesis consistent with the five core principles.

Several substantive and structural recommendations for follow-on to the EO tasks have emerged.  Here are two: BENS and CBA/CSCMP.  There will be more. Many of these operational recommendations are coherent with the five core principles identified above. Many of the approaches offered so far are government-centric.  Many of these recommendations would assist the US government play a more constructive role in Supply Chain Resilience.  But any government-centric role will, on its own, fail to engage many risks and opportunities for Supply Chain Resilience.  Demand and supply networks are not inherently governmental. In her 2009 Nobel Prize Lecture Elinor Ostrom said:

Most modern economic theory describes a world presided over by a government (not, significantly, by governments), and sees this world through the government’s eyes. The government is supposed to have the responsibility, the will and the power to restructure society in whatever way maximizes social welfare; like the US Cavalry in a good Western, the government stands ready to rush to the rescue whenever the market ‘fails’, and the economist’s job is to advise it on when and how to do so. Private individuals, in contrast, are credited with little or no ability to solve collective problems among themselves. This makes for a distorted view of some important economic and political issues. 

Such a distorted angle on reality will fail to advance Supply Chain Resilience. A stubborn adherence to this distorted angle actively threatens Supply Chain Resilience. In addition to the government reforms outlined by others, there is a need for a whole-of-nation approach to developing Supply Chain Resilience.  This would most likely be achieved through a Congressional chartering of a public-benefit corporation that would, consistent with the five core principles:

• Conduct data-gathering, relationship-building, analysis, and synthesis focused on the ecosystem level of demand and supply networks.
• Convene and facilitate meaningful and regular feedback regarding flows and fitness of demand and supply networks.
• Recommend and when possible facilitate voluntary mitigation and preparedness activities to advance Supply Chain Resilience