Exiting a Covid-19 lockdown: the bumpy road ahead for many supply chains

While the question asked is often how the market demand will recover, we believe it is much more important to understand what the supply chain response should be. Today, I have released the working paper “Exiting a Covid-19 lockdown: the bumpy road ahead for many supply chains”, that I have written over the past week with my colleague, former student, and friend Professor Maximiliano Udenio of the KU Leuven Research Center for Operations Management. In our paper, we argue that the best recovery strategy is actually quite independent of the exact timing and pattern of market recovery; it is much more important to know the current status of the supply chain in terms of inventory surpluses (or shortages) and cumulative lead-times. This especially holds for companies that are further upstream, such as those in the chemical or semiconductor industries. Similarly, those companies downstream should carefully manage their supply pipeline, as shortages may even appear if there are no further lockdowns of suppliers.

Supply chain theory tells us that the effects of sudden demand and supply shocks can be large. Anyone that has played the beer distribution game has experienced that even a small but sudden change in consumer demand may lead to large demand variations upstream. Such demand variations are exacerbated by the lack of information transparency and by long lead-times. This bullwhip effect can be observed in virtually every supply chain on a day-to-day basis.

While the Covid-19 crisis is unprecedented, and exact predictions are hard to make on how demand will unfold over the next few months, we believe it is possible by now to get some insights into the bumpy road ahead, and what companies could do to best ride these waves. We have done so by making use of models that we successfully used to predict the course of demand in the aftermath of the 2008 collapse of Lehman Brothers, leading to an extensive global credit crisis with major supply chain disruptions. In the paper we released today, we outline our initial findings on the ramp-up expectations after the Covid-19 lockdown exit.

Main insights to better manage recovery

Our main insights can be summarized as follows:

1. In many supply chains, especially the ones where at one or more echelons sourcing takes place from other continents or by components with long supply lead-times, dramatic inventories are currently building up, are unavoidable, and need to be properly managed in the next two months.
2. This inventory build-up will last for several months, causing near-zero orders for companies upstream in supply chains until inventories will start to decline due to recovered demand.
3. Inventories will start to decline by the end of the (Northern hemisphere) Summer of 2020 in most cases, implying that production for long-lead-time items would need to be started up ahead of this decline.
4. If this is not done (and this is highly likely for non-integrated supply chains without visibility), severe shortages of supply may start to build up in the late Summer and early Fall of 2020. This situation is very similar to the shortages that occurred in many supply chains in Spring 2009.
5. In markets where demand has been virtually eliminated over a longer period of time (due to market lockdowns), the effect of the supply lockdowns on the overall supply chain dynamics is small. In markets that have been less affected, the supply lockdowns and market demand reductions start interfering and lead to complex supply chain dynamics that are quite sensitive to the parameter estimates.
6. Companies that actively reduce their inventories in the coming months for whatever non-supply-chain reason, may shoot themselves severely in the foot; in all scenarios, shortages of supply are predicted in about 3-6 months.

Detailed graphs with multiple scenarios with a monthly 2020 timeline are included in the paper. See the bottom of this post for the download options.

While it is difficult to mitigate these effects, it is possible for companies to manage them more carefully, especially by actively managing the upstream and downstream supply chain.

The current crisis caused by the global Covid-19 pandemic has — so far — not been driven by an inventory adjustment but rather by a collapse in demand due to market lockdowns, and a collapse in supply due to production lockdowns. We know from supply chain theory that such shocks lead to oscillations across the entire supply chain. In our paper, we have tried to shed some light on these oscillations. It is common practice of researchers to model such oscillations such that also validation can be properly done, comparing the model predictions to the actual realizations. The exceptional circumstances of this crisis has encouraged us to share our model outcomes without ex-post validation. While this definitely does not complete a full empirical research cycle, we do believe our insights are valuable since we essentially extend out prior learnings that have undergone peer review.

Information-sharing and fact-finding is key to a build a competitive recovery strategy

Our analysis shows that — as expected — the demand and supply lockdowns lead to large fluctuations in orders across the supply chain, along with anticipated period of significant inventory surpluses and stockouts. Provided that recovery is steady (although maybe very slow), our models suggest that the main fluctuations should be dampened out by the middle of 2021. It should be noted this the relatively short period of instability is helped by the fact that in our settings there is a lot of smoothing taking place; if the smoothing is much less and companies respond quickly by drastically adjusting forecasts up and down, the instabilities are likely to be around much longer.

Perhaps less intuitive is that for markets where demand has almost been eliminated during the market lockdown, such as in the automotive industry, the effect of supply lockdowns on the dynamics is relatively minor: the market lockdown essentially determines the core of the dynamics. In that sense, it is very important to get a good insight in the market recovery. In the European automotive market, for instance, it is common for car makers to work with a customer order backlog of multiple months. They may enable to automakers to smoothen the effects of the sales elimination. However, even if this were the case, it is critical for the OEMs to communicate their production plans not just two weeks ahead (as is common in regular business), but 3-4 months ahead of time; this will enable also suppliers in 3rd, 4rd, and 5th tiers to better prepare. For markets where the reduction in demand has been much less, the effects of supply and demand lockdowns interact: the dynamics are more complicated with multiple shorter periods of increase and decrease in orders and inventory, and an increased imbalance between raw material/WIP and finished goods inventories, but the amplitudes are smaller and overall effects seem to be more moderated.

In all scenarios, we do observe a large inventory surplus currently building up. Anecdotal checking of this at multiple companies in various industries suggests this may indeed be happening. Also indexes about storage facilities being in high demand suggest this to be the case. Our results show that the inventory surpluses may be so high that companies far upstream in the supply chain may be stripped from any orders for months to come. This is mainly driven by having produced much more than market demand in the last few months, but in addition the bullwhip effect due to long lead-times and information asymmetry has further aggravated this effect.

What is now key for any company is to get a good insight into the cumulative inventory across the supply chain, as this will drive the supply chain’s dynamics over the next 6-8 months. Companies that manage to get such insights and act strategically in terms of managing their production and inventories will have the opportunity to outperform those companies that — driven by short-term financial objectives — make uninformed decisions to reduce their inventories.