Supply Chain Resilience

Price is lower than last year, but high enough that EU demand for natural gas continues much lower than two years ago– yet is sufficient to pull LNG flows from far away to replace Russia’s flows that are now mostly gone. Matching supply and demand depends a great deal on winter weather. Temperatures are forecast to be normal to slightly above normal and even “winter as a whole is predicted likely to be warmer than average…” Below is a very helpful pipeline flow map from S&P Global with further information on floating and other flows (access link for more details).

The most recent US focused Logistics Managers’ Index (LMI) offers a potential inflection point.

In August the Logistics Manager’s Index read in at 51.2. This is a marked change for the index as before August’s reading the overall index had registered three consecutive months of contraction and five consecutive months of registering new all-time low scores. In contrast, this is the fastest rate of expansion since February. The expansion this month is driven by increased activity across all eight sub-metrics of the index. Inventory Levels are still contracting, but at a much slower rate (+6.0) than July’s reading of 41.9, which was the steepest rate of contraction in the history of the index. This has led to an increase in Inventory Costs (+8.6 to 69.1) and Warehousing Prices (+2.8 to 63.4). We also observe Transportation Utilization moving out of contraction (+8.2 to 50.0) and the rate of Transportation Price contraction slowing considerably (+7.3 to 42.9).

Will this change in direction persist?

The more expansive Global Supply Chain Pressure Index (GSCPI) — that seemed to have bottomed out in May — has spent July and August idling at almost one standard deviation below long-term-average performance. Will pressure build or leak as we head into the Northern Hemisphere’s winter?

Especially when more comprehensive measures, such as the LMI and GSCPI, are not readily coherent, I find it helpful to follow some other measures. This spring and summer I have tried to tracking the following five:

North American Agricultural Production: The September 14 USDA Wheat Outlook reports:

U.S. Hard Red Winter (HRW) exports are forecast down 10 million bushels this month to 155 million bushels, the lowest since records started in 1973/74. HRW supplies have seen a long-term downturn in U.S. acreage as corn and soy have gained acreage in many locations. At the same time, international wheat competition has surged, resulting in exports of this class being less competitive on the global market. Recently, U.S. HRW supplies have been affected by drought in consecutive years, which has dented crop prospects and contributed to exports of this class being uncompetitive with other suppliers such as Russia and the European Union (EU). Historically, HRW was the leading class of U.S. exports, but in this season, it is forecast as the third largest class of U.S. exports, being surpassed by both Hard Red Spring (HRS) and White wheat (figure 1). Production of HRS and White are down year-over-year with lower yields, but drought has not affected those classes to the same extent as HRW.

The September 12 USDA World Agricultural Supply and Demand Estimates (WASDE) reported on the wheat crop, “Supplies are projected to decline 7.2 million tons to 1,054.5 million, primarily on lower production for Australia, Canada, Argentina, and the EU, which is only partly offset by an increase for Ukraine. If realized, this would be the first year-to-year decline in global wheat production since 2018/19. Australia is reduced 3.0 million tons to 26.0 million as dry weather this past month in Western Australia, New South Wales, and Queensland lowers yield prospects. Canada is decreased 2.0 million tons to 31.0 million on the initial model-based forecast by Statistics Canada for the 2023/24 crop, indicating lower yields from last year arising from dry conditions across the Prairies.” (More and more and more).

Global Natural Gas Demand and Supply: US domestic and export flows have been strong, at least prior to last week’s problems at the Freeport LNG terminal (more and more ). US exports combined with with outbound flows from Norway, Qatar, and Algeria have been sufficient to ameliorate recent concerns regarding Australia’s LNG exports. But Freeport’s troubles coinciding with Australia’s labor issues (and more) contributed to last week’s upward movement in the benchmark EU natural gas price. According to CNBC, Ana Maria Jaller-Makarewicz, energy analyst at the Institute for Energy Economics and Financial Analysis offered, “Gas markets are becoming riskier — gas and LNG prices are increasingly volatile and greatly affected by global factors… The uncertainty of future events that could affect gas supply makes it extremely difficult to predict how the supply and demand could be balanced and how much prices could escalate by. As seen in last year’s events in Europe, the only way that importing countries can mitigate that risk is by reducing their internal consumption.” In most of the EU demand and consumption of natural gas has been reduced by almost one-fifth multi-year averages. According to Reuters, “Gas imports into Germany dropped 17.9% year-on-year in January-August 2023…” Conservation has combined with a slowed economy to constrain natural gas consumption. In East Asia demand has been “tepid” for most of this year. US demand for natural gas has, however, been record-setting, largely due to electric power sector consumption in response to record summer heat.

China Export Volumes and Value: The Wall Street Journal reports that when valued in US dollars, “China’s outbound shipments declined 8.8% in August from a year earlier, China’s General Administration of Customs said Thursday [September 7]. The reading narrowed from the 14.5% year-over-year drop in exports in July, which marked the worst such result since February 2020.” (More and more). In July Mexico replaced China as the largest source of US imported goods. This reflects a sustained decline in US-China bilateral trade and shifts in China’s outbound velocity. The ASEAN region now claims more of total China trade (imports and exports) than either the United States or European Union. For August 2023 China’s dollar-valued exports to the US totaled $45,032 million, exports to ASEAN totaled $42,866 million, and exports to the EU $41,295 million. The South China Morning Post reports, “While China has long been Asean’s biggest trading partner, Asean only became China’s biggest trading partner in 2020 – filling a position long held by the US. Since 2018, when the trade war broke out between the world’s two biggest economies, changes in the supply chain have seen Beijing step up trade with its neighbours as manufacturers shifted from China to Asean countries. China-Asean bilateral trade grew from US$641.5 billion in 2019 to US$975.3 billion in 2022” (more and more). Bloomberg reports, “… parts sourced from China are increasingly moving to Southeast Asia for final assembly before being exported to the rest of the world…”

North American Grid Capacity: In early Spring this measure was selected because of concerns for getting through the Summer without major grid failures. We have had too many close calls (here and here and here), but the near-term worst is now probably behind us and longer-term trends are encouraging. The strategic risk involves 1) demand-spikes from extreme weather, especially in places with 2) rapidly increasing long-term core-demand where local generation capacity is challenged to grow fast enough, and 3) the ability when necessary to complement local generation capacity with effective transmission of wider-area — even continental — generating capacity. Wherever two or three of these factors gather together problems can (too) quickly trip into crisis. Extreme winter weather will increase prospects for this three-headed hydra. These risks are likely to persist for at least another decade in places with rapid population growth. But the North American grid is also experiencing significant capacity upgrades (here and here and here). But the transmission network to deliver this new capacity to demand is not expanding quite as quickly (here and here and here). The next few years will be dicey, but the problems are finally widely recognized, significant investments are being made, and — maybe — a shared sense of practical urgency has emerged (here and here and even here at this blog).

US Personal Consumption Expenditures: At the end of August the Bureau of Economic Analysis gave us this angle on July economic activity: personal income increased $45.0 billion (0.2 percent at a monthly rate). Disposable personal income (personal income less personal current taxes) increased $7.3 billion (less than 0.1 percent). Personal consumption expenditures (not inflation-adjusted) increased $144.6 billion (0.8 percent), see chart below. This level of spending generates strong corporate balance sheets, but over time has the opposite effect on personal financial conditions. Pandemic-related forced savings and wage increases have fueled several months of increased personal expenditures but the surplus is almost gone (here and here). How quickly and how much this pull will shift is now a crucial question. Will wages leap higher? If so, how much will inflation eat away the increase? Or how many jobs will be lost as interest rates rise to dampen inflation? If fuel costs continue to increase (see above) this will increase costs and prices across sectors. How will PCE adapt? Who will win? Who will lose? How will push respond to these changes in pull?

With many local exceptions, it seems to me that push and pull have been basically in balance for most of this year. Despite droughts, war, and more, food remains abundant — if a bit less abundant than in some prior years. For most of the last year global energy stocks have also been abundant, more abundant than reasonably feared, and prices have been moderate. There are now increasing pressures — some intentional and some not at all — to constrain energy flows in a manner that tends to increase prices. Marginally increased prices for fuel, food, and more have, so far, softened demand more successfully than spurred increased wages. In the late-pandemic period and since, European and East Asian consumers have been restrained. Even US consumers have reduced their rate of increased spending when adjusted for inflation (see red line in chart below). Upstream production capacity has mostly been preserved (and even expanded), but actual push has slowed to better match current pull. This balance can be treacherous in case of any sudden, unexpected spike in demand or when midstream channels are insufficient (e.g., Panama Canal or the grid’s transmission network) or when a significant upstream source is taken offline (e.g., LNG terminals or a failed harvest).

How this balance might shift in the final quarter of 2023 is not obvious to me. I can make contending arguments. But in terms of current fitness, I perceive that global supply chains are in healthy shape for the start of the Northern Hemisphere’s winter.

Overnight the United Auto Workers initiated walk-outs at assembly plants operated by Ford in Wayne, Michigan by GM in Wentzville, Missouri and by Stellantis’ Jeep brand in Toledo, Ohio (more and more). The walk-outs involve 12,700 workers out of a total UAW Big 3 membership of almost 150,000. According to Reuters, this job action “will halt production of the Ford Bronco, Jeep Wrangler and Chevrolet Colorado pickup truck, along with other popular models.”

This is a surgical strike intended to maximize Big 3 pain while minimizing the proportion of UAW members hurt (and draw-downs on the UAW strike fund.) The vehicles no longer being produced are among the most profitable in each company’s line-up. This reflects a sophisticated understanding of the auto industry’s value-chain as well as its supply chain.

As this blog often notes, contemporary high volume, high velocity (and high value) supply chains tend to be highly concentrated. When these capacity concentrations are hit hard — intentionally or not — there is an amplified effect. Supply Chain Resilience attempts to identify these “targets” and reduce systemic vulnerability. It appears that the UAW has conducted similar analyses to exploit Big 3 vulnerabilities in a way that requires the least cost to the UAW.

Much smarter than any hurricane.

+++

September 19, Update: Bloomberg has a Big Take on the strike entitled, How Auto Executives Misread the UAW Ahead of Historic Strike. The story highlights personal tensions between UAW President Shawn Fain and auto industry leaders. The Bloomberg narrative prompted bad old memories.

From late 1979 into early 1980, I was part of a crisis communications team consulting with International Harvester — then among the top ten largest US companies — as it “negotiated” a new contract with the United Auto Workers. I was not involved with IH prior to the UAW beginning strike action, but it is my impression that IH senior management anticipated, even welcomed, the strike as a means to wrangle substantial salary concessions from their employees.

Archie McCardle, CEO of International Harvester, was two years into a major restructuring of the company. He envisioned a radical reordering of corporate objectives and behavior calibrated to increasing global competition and post-OPEC economic realities. Especially in retrospect, it was a smart strategy. McCardle and the Board were working to proactively manage seismic changes that did reshape the global economy during the last quarter of the Twentieth Century.

Many labor-management structures that made sense in the 1960s were unsustainable by the 1980s. But while most US corporations (and even some labor unions) recognized this reality, cost-containment and incremental streamlining became the typical transition path. Instead McCardle insisted on radical, arguably rational reductions in labor costs and much more agility to deploy labor assets.

McCardle came to see the UAW as an existential threat to his strategic vision. I’m not sure this was his position at the beginning of the strike, but by Christmas 1979 McCardle had become a wanna-be Union Buster. He despised Union leaders. The feeling was soon mutual. In meeting after meeting I watched predictable — thereby manageable — disagreements descend into fever-dreams of mutual destruction. I also came to perceive my client, Mr. McCardle in particular, as the principal aggressor. There was bad behavior on all sides. There were also opportunities to defuse the conflict that Mr. McCardle used to escalate tensions. I resigned my modest role sometime in late February.

The cascading financial consequences of the 172-day strike essentially destroyed International Harvester… and the lives of many employees.

This is not — thank goodness — a perfect analogy for the current UAW strike. The Big 3 have offered more not less. Mr. McCardle was an unusually abundant source of arrogance, disdain, and aggression. Yet I expect arrogance is at play today, the Bloomberg piece offers some evidence. I am certain the automobile industry — and the global economy — is in the midst of a fundamental inflection point. These transitions are always treacherous. When any of us feel under attack, rationality can be an early casualty.

Supply chains are sometimes called socio-technical systems. Vast flows of digital signals now communicate precise measures of demand velocity. Pallets are planned by algorithms and sometimes picked by robots. In the last half-century technological change has fundamentally altered the relationship of demand and supply. The technologies of the last thirty years have accelerated this revolutionary change. Today’s technology is supercharging change.

But again and again those of us inside the system encounter the deeply human, profoundly social character of the network. Every minute choices made by just one person speed or impede flow. Our technologies tend to be linear, while reality is usually not. Human relationships inform and shape flows in ways that, so far, our technologies do not anticipate or stupidly over-estimate. CEOs meet and do deals when personalities mesh as well as their spreadsheets. What looks great on paper explodes when personalities clash. Progress is often the product of mutual restraint.

Many of us have been on a dock or in a fulfillment center or on a production line that is humming happily and experienced the ugly results when that hum is replaced by congestion, collisions, shouts, and excuses. In either case technology is seldom the principal cause. Technology usually amplifies social content and context.

Today the UAW and the Big 3 are engaged in deciding how a big part of our world will behave going forward. This task is tough enough even if union and company leadership behave themselves. The more they behave badly, the more likely bad results for everyone.

September 26 Update: Late last week Ben Cohen at the Wall Street Journal explained how the current UAW strike strategy is related to a 1990s strike by flight attendants against Alaska Airlines. The playbook for that labor action was named Create Havoc Around Our System (Chaos). Great acronym. Very accurate use of “system” (from Ancient Greek σύστημα (sústēma, “musical scale; organized body; whole made of several parts or members”), from σῠνίστημῐ (sunístēmi, “to combine, organize”) + -μᾰ (-ma, resultative suffix). σῠνίστημῐ is from σῠν- (sun-, “with, together”) + ἵστημι (hístēmi, “to stand”), from Proto-Indo-European *steh₂- (“to stand (up)”)

The UAW’s labor action — like the Association of Flight Attendants’ approach before it — is targeting core characteristics of each sector’s system service levels. Cohen wrote, “Both are meant to sow confusion, keep companies guessing and paralyze interconnected systems. Both amplify uncertainty and create opportunities for mistakes that the union can exploit.”

By intentionally — strategically — striking some places and not others a comparably few strikers can have a cascading systemic effect. The UAW’s recent escalation of strike action against General Motors and Stellantis (but not Ford) increased the number of strike venues and shifted from upstream vehicle manufacturing to midstream parts distribution (more and more and more). Hits to the midstream will accelerate downstream impacts and distress.

Airlines are even more networked than automobiles, but especially in the last twenty years the US automotive industry has sufficiently streamlined and concentrated its network topology (more) that it is much more efficient on most days and much more vulnerable to hard-hits in the wrong-places on bad days.

The UAW calls this their “Stand Up Strike.” This is intended to recall successful “Sit Down Strikes” of the 1930s. It is also — no doubt unintentionally — an etymologically accurate description of intentional targeting of capacity concentrations.

Below are multi-year slopes for the Consumer Price Index for food. Here is the complete Bureau of Labor Statistics’ explanation of what happened in August:

The food index rose 0.2 percent in August, as it did in the previous month. The index for food at home increased 0.2 percent over the month, after rising 0.3 percent in July. The six major grocery store food group indexes were split over the month, with three increases and three decreases. The index for meats, poultry, fish, and eggs rose 0.8 percent in August as the index for pork increased 2.2 percent. The index for other food at home increased 0.2 percent over the month and the index for cereals and bakery products rose 0.5 percent. The index for dairy and related products decreased 0.4 percent in August after increasing 0.5 percent in July. The fruit and vegetables index declined 0.2 percent over the month, as did the nonalcoholic beverages index. The food away from home index rose 0.3 percent in August. The index for limited service meals rose 0.3 percent over the month, and the index for full service meals increased 0.2 percent. The food at home index rose 3.0 percent over the last 12 months. The index for cereals and bakery products rose 6.0 percent over the 12 months ending in August. The meats, poultry, fish, and eggs index was unchanged over the year. The remaining major grocery store food groups posted increases ranging from 0.3 percent (dairy and related products) to 4.8 percent (nonalcoholic beverages). The index for food away from home rose 6.5 percent over the last year. The index for limited service meals rose 6.7 percent over the last 12 months, and the index for full service meals rose 5.2 percent over the same period.

During most of 2020, 2021 and the first half of 2022 demand and supply for food (and much more) was totally discombobulated. Today’s nominal prices are often much higher than pre-pandemic and a serious gap emerged between price increases and wage increases. This was a source of considerable demand volatility to which supply chains were hard-pressed to adapt. During this period, food became much less affordable for the two least affluent economic quintiles.

But from a Supply Chain Resilience perspective, I perceive that since the beginning of 2023 the CPI for food — especially Food At Home — has been modest and healthy. This sort of nominal growth supports supply capacity and, given this year’s inflation-adjusted wage increases, are within reach of most households. The August CPI indicates some serious inflationary pressures (especially related to fuel and more), but 2023 grocery pull patterns are, so-far, well-matched to sustainable push capacity.

+++

September 14 Update: According to today’s Census Bureau report on August retail sales, seasonally adjusted sales at grocery stores increased from $73,644 million in July to $73,933 million in August (both are nominal, non-inflation adjusted figures). Given the population size, amounts of money involved, and inflation, this is about as flat as I would ever expect to see. Here’s the Fed’s related chart. Year-Over-Year there was much more spending growth in Food-Away-From-Home (8.5 percent compared to 2.1 percent), but the recent rate of growth in FAFH appears to be slowing as well.

As discussed in a recent post, Force On Target is often helpful in thinking through Supply Chain Resilience. In the case of Hurricane Idalia the Force that hit Florida, Georgia, and the Carolinas was more fierce than forecast two days prior to landfall. Rather than a Category 1 or 2 hurricane, it reached CAT4 and came ashore as a CAT3. The forecast track was remarkably consistent, so anticipated targets were by-in-large the actual targets.

Given the track, my principal Supply Chain Resilience worry was potential impacts of storm surge at Tampa Bay and especially the fuel infrastructure supporting local racks and the Central Florida Pipeline. If Tropical Force Winds (see first map below) and predicted storm surge had breached sea walls and port-side structures the consequences could be wide-spread, time-extended, and deeply troublesome. So… on Wednesday about 0400 local when the ocean buoy nearest that fuel infrastructure showed that the bay had moved past high-tide with a survivable peak (see chart below), there was palpable relief.

Beyond Tampa Bay my next priority concern was direct population impacts — and issues with demand-signaling — at and near landfall. As the Force factor increased my Target zone did as well, extending into Southern Georgia. A colleague did a great job of sizing the space and population that would be the principal target. Please see the two maps below.

The much deeper (if narrower) target zone on the second map is not dense with supply chain capacity. It is, in fact, almost a “demand desert” characterized by sparse population and high-proportional poverty (here is the census summary for Taylor County, site of landfall). The potential for this already anemic demand to be further isolated was a real concern as Idalia tore across rural areas of the Southeast United States. Especially given the potential for delayed indicators of increased isolation, it will be another day or so before I can confidently assess the ability of dispersed demand to access supply that has mostly recovered, but remains at least as distant as before Idalia arrived.