Realizing supply chain agility under time pressure: Ad hoc supply chains during the COVID‐19 pandemic

Abstract

When the COVID‐19 pandemic began in 2020, the medical product industry faced an unusual demand shock for personal protective equipment (PPE), including face masks, face shields, disinfectants, and gowns. Companies from various industries responded to the urgent need for these potentially life‐saving products by adopting ad hoc supply chains in an exceptionally short time: They found new suppliers, developed the products, ramped‐up production, and distributed to new customers within weeks or even days. We define these supply chains as ad hoc supply chains that are built for a specific need, an immediate need, and a time‐limited need. By leveraging a unique sampling, we examined how companies realize supply chain agility when building ad hoc supply chains. We develop an emergent theoretical model that proposes dynamic capabilities to enable companies building ad hoc supply chains in response to a specific need, moderated by an entrepreneurial orientation allowing firms to leverage dynamic capabilities at short notice and a temporary orientation that increases a company's focus on exploiting the short‐term opportunity of ad hoc supply chains.

1. INTRODUCTION

In 2020, the global outbreak of the COVID‐19 pandemic caused a surge in demand for essential personal protective equipment (PPE), since the virus was easily transmitted from person to person through air‐borne droplets (Johns Hopkins Medicine, 2021). As well as face shields, gowns, and disinfectants, a wide variety of types of masks was needed, including surgical masks, community (non‐surgical) masks, and respirator masks (i.e., U.S. Standard N95/N99 masks and European standard FFP2/FFP3 masks). All these types of PPE were required immediately to prevent infection between frontline workers who were directly exposed to the virus when treating infected patients or conducting COVID‐19 tests (Popovich & Parshina‐Kottas, 2020), while individuals were increasingly urged to wear face masks in public (Ferré‐Sadurní & Cramer, 2020).

The urgent need for massive volumes of PPE caused extreme time pressure to increase the supply of products that could determine the health of millions of people. Anticipating this need, the response from many manufacturing companies from industries other than those that traditionally produce PPE was remarkable: They moved swiftly beyond their core competences to build new supply chains for PPE. For example, the Chinese automotive manufacturer BYD opened a plant for face masks and disinfectants (BYD, 2020); the French spirits manufacturer Pernod Ricard repurposed distilleries to produce hand sanitizers at one of their U.S. production sites (Buckley, 2020); the Taiwanese electronics manufacturer Foxconn ramped up production of surgical masks (Davies, 2020); and the Canadian sports equipment manufacturer Bauer shifted its production to face shields (Waldstein, 2020).

We refer to supply chains built under time pressure as ad hoc supply chains, that is, those built in response to a specific need, an immediate need, and a time‐limited need. Under considerable time pressure, a number of companies demonstrated high supply chain agility, enabling them to construct ad hoc supply chains from scratch within an extremely short space of time. Within weeks or, in some cases, days, these companies sourced raw materials and production machines, designed new products, and developed a previously unknown market.

This study explores the phenomenon of ad hoc supply chains. Specifically, we ask how companies realize supply chain agility when building ad hoc supply chains under time pressure. Given the lack of prior research on ad hoc supply chains, we designed a multiple case study (Eisenhardt, 1989) using grounded methods to examine 34 German companies that adopted ad hoc supply chains for PPE during the COVID‐19 pandemic. These companies came from very different industries, were of different sizes, and had no prior experience in the PPE market. In two interview rounds in 2020, we conducted semi‐structured interviews with the companies' executives, key informants who planned and operated ad hoc supply chains. Following a structured coding procedure (Charmaz, 2006; Gioia et al., 2013), we subsequently developed a theory on ad hoc supply chains.

The central contribution of our study is to provide an initial theoretical understanding of ad hoc supply chains that are built for meeting a specific, immediate, and short‐term demand for products. We further inductively develop an emergent theoretical model of ad hoc supply chains including enablers of supply chain agility: dynamic capabilities (recognizing and responding to a specific need), an entrepreneurial orientation (developing ad hoc supply chains in short order) and a temporary orientation (speedy action within a limited time period).

In advancing the theoretical understanding of ad hoc supply chains, we contribute to several literature streams. Whereas the extant literature assumes that dynamic capabilities and ad hoc activities are unrelated constructs, we find that adopting ad hoc problem‐solving relies on dynamic capabilities. Our contribution to the entrepreneurship literature is to suggest that, under time pressure, market innovations are not compatible with the speed needed to build ad hoc supply chains. Finally, we introduce the construct of temporary orientation to the strategic orientation literature, describing company behavior that results in exploiting a short‐term opportunity.

The remainder of this paper is structured as follows. Section 2 outlines the literature on supply chain agility in line with our inductive research approach. Section 3 provides detailed background information about the COVID‐19 pandemic in Germany to establish the context for this study. In Section 4, we explain our methodological approach. We present our findings from the data analysis in Section 5. Based on our findings, in Section 6, we develop the theoretical model and propositions. In Section 7, we discuss the model and in Section 8 we conclude by summarizing our findings and providing suggestions for further research.

2. THEORETICAL BACKGROUND

Following an inductive approach, we studied the phenomenon of ad hoc supply chains independently from the existing literature. This approach acknowledges Eisenhardt's (1989) inductive research process, “where researchers walk in the door and don't have a preconception of what relationships they're going to see…[they] may have a guess about the constructs, but are fundamentally going in open‐minded” (Gehman et al., 2018, p. 287). We did approach the study with the suggestion that supply chain agility accounted for the speed realized by companies in response to the demand for PPE.

Supply chain agility refers to the ability of a firm and its supply chain partners to respond rapidly to uncertainties (Braunscheidel & Suresh, 2009; Lee, 2004; Swafford et al., 2006), due either to the risk of changes in demand (Braunscheidel & Suresh, 2009; Christopher, 2000; Gligor, 2014; Gunasekaran & Yusuf, 2002; Ismail & Sharifi, 2006; Lin et al., 2006; Sharp et al., 1999) or to the risk of supply chain disruptions (Braunscheidel & Suresh, 2009; Charles et al., 2010; Kleindorfer & Saad, 2005; Shekarian et al., 2020; Udenio et al., 2018; Yang, 2014). In our study, we refer to supply chain agility in the context of the disruption of global supply chains that resulted in a worldwide shortage of PPE. By demonstrating a high level of supply chain agility, many manufacturing companies responded to this global need by building ad hoc supply chains within a short time.

Building ad hoc supply chains under time pressure has been significant in past crises. For example, during World War II, manufacturers outside the defense industry devised supply chains for war materials at extreme speed. In the United States, lingerie companies sewed parachutes, lipstick makers produced bomb fuses (Mihm, 2020), and the automotive manufacturer Ford built B‐24 bombers (Ford, 2019). More recent instances of diseases and epidemics provide further examples: During the Ebola epidemic in 2014, humanitarian organizations constructed central warehouses from scratch within a couple of weeks (Logistics Cluster, 2014) and sourced suitable means of transport to obtain access to the disaster areas (Médecins Sans Frontières, 2020). Ad hoc supply chains have been historically significant for increasing the supply of urgently needed products and remain significant today given the numerous supply chain uncertainties (e.g., military conflicts, natural disasters, and trade wars) that we currently face.

3. RESEARCH CONTEXT

To contextualize our research temporally, in Figure 1 we outline the course of the COVID‐19 pandemic in Germany and the events relevant for companies building ad hoc supply chains. In Germany, the first wave of COVID‐19 was observed in the spring of 2020, with the first recorded infection on January 27. ¹ Thereafter, the number of new infections recorded daily increased steadily, reaching 6,000 in April. ^{2 , 3} To curb the spread of the virus, on March 16, the government announced the first national lockdown, closing schools, shops, and restaurants, ⁴ directly followed by travel restrictions for tourists, ⁵ and contact restriction regulations. ⁶ After the lockdown was lifted on April 15, 2020, ⁷ all 16 federal states agreed on April 29, 2020 to make the wearing of masks in public obligatory to prevent a new peak in infection. ⁸

FIGURE 1.

Timeline of COVID‐19 events in Germany, January–August 2020

Like most other countries affected by COVID‐19, Germany experienced very limited availability of PPE at the beginning of the first phase of the pandemic. To increase the PPE supply, the German government contracted more than 700 companies to deliver PPE by the end of April 2020. ⁹ As well as this initiative toward addressing short‐term supply, in May the government entered framework contracts with around 50 German manufacturing companies to deliver a fixed weekly quantity of respirator masks, surgical masks, and gowns, all to be produced in Germany. ¹⁰ To further increase the supply of PPE, the government encouraged local production by temporarily adjusting and relaxing manufacturing requirements and regulations for face masks and disinfectants. Specifically, nonmedical manufacturers were allowed to produce disinfectants. ¹¹ For face masks, a fast‐track certification procedure was introduced that allowed companies to skip the time‐consuming certification process until August 31. ¹² In addition to these short‐term regulatory relaxations, newcomers were given 30% co‐funding until June 30 for investments in machinery to produce surgical and respirator masks and meltblown materials for mask filters. ¹³ While the PPE shortages in the first wave of the infection were extreme, PPE was more readily available in the second infection wave that began in September 2020. Consequently, local PPE production incentives were withdrawn after the first infection wave.

4. METHODOLOGY

We designed a multiple case study to inductively explore the supply chain agility of companies building ad hoc supply chains (Eisenhardt, 1989). The case study approach is especially valuable to understand new and emerging phenomena for which no empirical evidence or theory is yet available (Eisenhardt, 2021; Eisenhardt & Graebner, 2007). We consider our research to be an important first step in the theoretical understanding of ad hoc supply chains. Our data allow us to conduct a highly exploratory approach that identifies enablers of adopting ad hoc supply chains. Nonetheless, further scientific work is needed to test and extend our theoretical model and propositions to exhaustively explain the interaction of the individual constructs that affect the adoption of ad hoc supply chains.

We identified more than 200 potential case study candidates among companies in Germany building ad hoc supply chains during the COVID‐19 pandemic, from which we constructed a unique sample of 34 (see Table 1). These companies had no previous experience with PPE and deviated from their core businesses when starting PPE production. The sample comprises companies of different sizes (e.g., start‐ups, medium‐sized, and multinational companies), and from different industries (e.g., textile, automotive, and chemical), all creating supply chains for different products of PPE (e.g., community masks, respirator masks, medical masks, disinfectants, face shields, and gowns). The companies also took on different supply chain functions, for example, sourcing materials, production, or distribution, which enabled us to gain insights from different perspectives (Eisenhardt & Graebner, 2007). Moreover, we included three established healthcare companies in our sample to understand the dynamics of the PPE market during the COVID‐19 pandemic from the expert's point of view (see Table 2).

TABLE 1.

Company sample characteristics

Company ID	Industry	Core business a	Sales in core business (in € million)	PPE type	Position of interviewee 1 (+interviewee 2 b )	First interview (April–June 2020)	Second interview (November 2020)
CO1	Textile	Clothing	101–500	Community masks, respirator masks	Chief operating officer	X
CO2	Textile	Nightwear	11–50	Community masks	Chief executive officer	X	X
CO3	Textile	Home textiles	11–50	Community masks	Sales manager	X	X
CO4	Pharma	Tablets and capsules	51–100	Disinfectants	Production director	X	X
CO5	Fashion	Underwear	51–100	Community masks	Chief executive officer	X
CO6	Chemical	Stains and paints	51–100	Disinfectants	Chief executive officer	X
CO7	Fashion	Luxury clothing	1–10	Community masks	Chief executive officer	X
CO8	Chemical	Water chemicals	101–500	Disinfectants	Chief operating officer	X	X
CO9	Fashion	Luxury clothing	1–10	Community masks	Chief executive officer	X
CO10	Textile	Filter textiles	1–10	Respirator masks	Sales manager	X
CO11	Medical textile products	Orthoses	51–100	Respirator masks	Head of R&D + consultant	X	X
					Procurement manager	X
CO12	Paper	Artist and filter papers	11–50	Paper test strips, community masks	Chief executive officer	X
CO13	Chemical	Chemicals	51,000–100,000	Disinfectants	Senior vice president logistics	X
					Vice president logistics	X	X
CO14	Chemical	Specialty chemicals	5001–10,000	Disinfectants	Head of operations	X
CO15	Fashion	Underwear	1000–5000	Surgical masks	Head of supply chain management	X
CO16	Automotive	Automotive filter parts	10–50	Meltblown material supplier	Head of filters	X
CO17	Beverage	Liquor	1–10	Disinfectants	Chief executive officer	X
CO18	Fashion	Shirts and blouses	101–500	Community masks	Chief operations manager	X
CO19	Engineering	Tool engineering	11–50	Mask production machines	Chief executive officer	X
CO20	Packaging	Packaging solutions	5001–10,000	Gowns	Plant manager + communication manager	X
CO21	Automotive	Automotive interior equipment	10–50	Surgical masks, respirator masks	Head of PPE	X	X
CO22	Chemical	Pharmaceuticals	501–1000	Disinfectants	Operations manager	X
CO23	Automotive	Acoustics and insulation	10–50	Respirator masks	Chief executive officer	X	X
CO24	Healthcare	Medical products	51–100	Sales of surgical masks and gowns	Project manager	X	X
CO25	Healthcare	Medical products	51–100	Gowns	Chief executive officer	X
CO26	Automotive	Technical textiles	1–10	Surgical masks	Chief executive officer	X
CO27	Textile	Fashion, technical, and home textiles	1–10	Surgical masks	Chief executive officer	X
CO28	Healthcare	None	Start‐up	Surgical masks	Founder 1 + Founder 2	X	X
CO29	Packaging	Product packaging and advertising	11–50	Face shields	Chief executive officer	X
CO30	Engineering	Machine and plant engineering	1001‐5000	Surgical masks	Managing director	X	X
CO31	Automotive	Light components	51–100	Community masks	Chief executive officer + head of product development	X
CO32	Healthcare	Medical devices and hygiene products	501–1000	Surgical masks	Chief operations manager	X
CO33	Engineering	Heating and cooling systems	1001‐5000	Ventilators	Chief technology officer	X
CO34	Healthcare	Pharmaceuticals	10,001–50,000	Disinfectants	General manager	X
					Head of supply chain management	X

^a The term core business is not valid for the sampled start‐up.

^b For the cases that the first interviewee was joined by a colleague in the interview.

TABLE 2.

Expert sample characteristics

Expert ID	Industry	Core business	Sales in core business (in € million)	Position of interviewee 1 (+ interviewee 2 a )	First interview (April–June 2020)	Second interview (November 2020)
EXP1	Healthcare	Manufacturing medical technology	10,001–50,000	Strategic manager	X
EXP2	Healthcare	Manufacturing healthcare products	1000‐5000	Sales manager	X
EXP3	Chemical	Manufacturing disinfectants	101–500	Head of supply chain management + strategy manager	X	X

^a For the cases that the first interviewee was joined by a colleague in the interview.

4.1. Data collection

Our study draws on interview data collected during the COVID‐19 pandemic. We conducted semi‐structured interviews for each case company between April and June 2020. In this time period, the sample companies adopted ad hoc supply chains, reducing the potential for retrospective bias in the data (Eisenhardt & Graebner, 2007; Voss et al., 2002). We interviewed executives who managed the ad hoc supply chain project, and so were highly knowledgeable about how to realize supply chain agility (Eisenhardt & Graebner, 2007). While we focused on conducting interviews with individuals who played leading project roles, in some cases we talked to two interviewees simultaneously, that is, for CO11, CO20, CO28, CO31, and EXP3. Furthermore, for companies CO11, CO13, and CO34 we talked to two interviewees separately to gain more specific insight into the project.

The semi‐structured interview guideline had three parts (see Appendix A1) and included open questions, not derived from existing literature, to reflect the exploratory element of our research. In the first part, we asked our interviewees general questions about their ad hoc supply chain, that is, the PPE product, the internal and external resources required (expertise, machinery, materials, and suppliers), production planning, sales, and distribution. In the second part, we focused on the company's specific challenges, such as certification and material sourcing, followed by questions about their learning. While we made sure we covered each question in the interview, we allowed the interviewees to structure the conversation by not specifying the order of the questions.

We conducted the interviews in pairs, which allowed us to discuss the insights after each interview. As interesting aspects arose, they were formulated as questions in the guidelines for subsequent interviews. Topics of less theoretical interest were not addressed in later interviews. Due to the travel and contact restrictions imposed during the COVID‐19 pandemic, we talked to the interviewees by phone or video. The average length of interviews was 39 min. Since the interviewees were generally German native speakers, we conducted the interviews in German, except for one interview with an English native speaker. Each interview was recorded; however, one interview relied exclusively on written notes due to technical issues. All the interviewees except one signed a consent form giving us permission to process the data collected for this study. The interviewee who refused permission was excluded from the analysis. The recorded files were transcribed by a third party and resulted in 613 pages of interview transcript.

We approached a subset of these companies for follow‐up interviews in November 2020, minimizing the potential risk of impression bias (Eisenhardt & Graebner, 2007). In this second round, the average length of interviews was 26 minutes. Our interview guideline included open questions about the status quo of PPE production, learnings from the speed realized during the project, retrospective evaluation of the projects, and the outlook for the future of the PPE business, should it be integrated into the core business (see Appendix A2).

As well as manufacturers that had started PPE production, we interviewed three experts from established healthcare manufacturing companies in both rounds of interviews. We put industry‐related questions to them to understand developments in the healthcare industry at this time, evaluation of the realized supply chain agility, learnings from the crisis, and the outlook for local PPE supply chains post‐crisis (see Appendices A3 and A4).

4.2. Data analysis and theory‐building

We conducted a structured coding analysis and a replication analysis to develop a theory on how companies realize supply chain agility when adopting ad hoc supply chains. We identified theoretical constructs for the coding analysis in two phases; in line with our inductive approach, we did not apply prior code structures derived from theory. Instead, we allowed topics to emerge from the data so that our findings were not limited to existing knowledge (Gehman et al., 2018). In the first coding phase, we coded the raw data of each interview with initial codes (Charmaz, 2006, pp. 45–48), often verbatim or paraphrasing, to represent as closely as possible the information provided by the interviewee. During this initial coding phase, we translated the German‐language raw data to English‐language initial codes. From this point, we continued to create codes in English.

In the second coding phase, we structured the initial codes to derive theory from the data, similar to the approach introduced by Gioia et al. (2013). We categorized the codes into four levels of increasing abstraction to develop theoretical constructs, grounded in the data (Eisenhardt, 2021). While developing the code structure, we consulted prior literature to see whether our codes represent existing theoretical constructs (Eisenhardt, 1989). By cycling between our data, emergent theory, and existing theory, we refined the code structure in several iterations until we arrived at the final code structure presented in the Online Appendix I.

After developing the code structure, we conducted a replication analysis (see Online Appendix II), which allowed us to analyze whether emergent concepts were confirmed across the cases (Eisenhardt, 1989; Eisenhardt & Graebner, 2007; Yin, 2018). In Online Appendix II, cases that provide evidence for a specific concept are marked X, whereas cases that do not provide sufficient support for the concept are marked ‐. Although the concepts are not evident across all cases, the data “show a close fit” (Eisenhardt, 1989, p. 541) with the identified concepts, which allows us to build a good, if emergent, theory (Eisenhardt, 2021).

Through the structured coding process, we identified several constructs that enable companies to adopt ad hoc supply chains. We developed propositions as to how these constructs are linked and visualized their relationships in a theoretical model in accordance with Eisenhardt's definition of a theory: “A theory is a combination of constructs, relationships between constructs, and the underlying logic linking those constructs that are focused on explaining some phenomenon in a general way” (Gehman et al., 2018, p. 291).

5. HOW COMPANIES REALIZE SUPPLY CHAIN AGILITY WHEN BUILDING AD HOC SUPPLY CHAINS

To build theory on how companies realize supply chain agility when creating ad hoc supply chains, we first present the findings of our data analysis, which identified five theoretical constructs described in‐depth in the following. The constructs are empirically supported by a high replication rate across the case companies, shown in Online Appendix II. Online Appendix III provides an overview of these constructs, their descriptions, and supporting literature. Table 3 summarizes our findings and provides direct quotes that represent the concepts identified.

TABLE 3.

Representative quotes

Concept	Representative quote
Ad hoc supply chains
Specific need	“We have to survive somehow. And within this phase, we then thought, or I thought at the end, what can we do to make this as good as possible, to somehow persevere and, perhaps still provide a service to society?” (CO11)
Immediate need	“And have had basically nothing to do with masks at all until mid‐March, about March 16, and then switched to face masks.” (CO2)
Time‐limited need	“And now these quantities [from Asia] are also coming into the market at very, very favorable prices. That is why we can see that our business will now come to an end relatively quickly, for price reasons. […] This has been a good interim solution for us now, but that is going to be difficult to sustain in the long term.” (CO5)
Supply chain agility
Time pressure	“So, right now, the [customer] who just called me, he needs 10 tons of disinfectants, preferably tomorrow.” (CO14)
Extreme speed	“So, a process like that, I think it would usually take at least half a year to a year of preparation. Here, we realized everything in two or three weeks. “(CO11)
Dynamic capabilities
Internal capabilities
Leveraging knowledge	“Producing masks is about cutting and ultrasonic welding, and those are processes that we do here in various forms in‐house and at our other sites. At the end of the day, that is our expertise in making textiles.” (CO27)
Exploiting capacities	“We knew that we do not really need to worry about demand but rather about capacity. And that is why we tried to increase production capacity to its maximum.”(CO1)
Engaged employees	“That is the willingness of the team to go the extra mile. The fact that none of us leaves work before normal working hours is not questioned by anyone. That is a great team effort.” (CO21)
Leadership skills	“I had to do much convincing. That was also extremely important to get some people on board at the beginning. Then, I really picked up one [employee] after another. And the ones I picked up, organized the whole thing.” (CO1)
External capabilities
Activating networks	“If you really do buy on all continents, have the networks, know suppliers, or have access to a network of suppliers that you may not have known in the past but now urgently need, then you can find these contacts.” (CO13)
Trusted collaboration	“We do not have the capacity to produce 300,000 masks per week ourselves. That is why we have been working based on a contract manufacturing concept right from the start. You basically provide the complete material and the work instructions, and then, other companies sew for you.” (CO11)
Supply chain reliability	“We received an order for 1.8 million respirator masks from the federal state authorities. We also received a letter of intent very early. Without the LoI, at the least, we probably would have had a hard time getting the goods.” (CO11)
Entrepreneurial orientation
Proactiveness	“We ended up having a lot of components on stock and it just took this: We can do this! Let us do this! And then be fast. That was ultimately the essential leap.” (CO22)
Risk‐taking	“We have made the investments from our own funds so far and from external funds for the automation. So, there are several million that we will put into production here.”(CO21)
Competitive aggressiveness	“So there were three products, but now there are considerably more. In the beginning, we did not give the customer a choice. But in the meantime, he can choose fabrics, and he can also choose different types. For example, we now have three new types of fabric masks and we are starting with a children's mask.” (CO1)
Temporary orientation
Pragmatic problem‐solving	“As nose clips, we still use bread clips. But that is not at all unusual because they are easy to form. There is a double wire in there that is overmolded with plastic, so it is from bread packaging or pasta packaging or something like that.” (CO11)
Process alignment	“The challenge is also planning complexity. So if you have to react very quickly, we immediately realized that we will produce a mask that is not a standard product for us, that was not created in the system, in the ERP system. And that is why we used Excel to help us out.” (CO5)
Geographical proximity	“We would not have looked for a cheap label supplier from India. Of course, we have taken the one here around the corner, which is a bit more expensive but slightly faster. An ad hoc supply chain is more local than global.” (CO22)
Assembling project teams	“But, of course, there are a few keys to success that are essential. You set it up as a task force and give the people all the freedom possible. So, in principle, they do not need to follow any company rules at all.” (CO33)

5.1. Adopting ad hoc supply chains

Grounded in the data, we developed the new construct of ad hoc supply chains, which, we argue, differ from regular supply chains in core businesses in terms of the specific, immediate and time‐limited need they fill. First, ad hoc supply chains are adopted for the purpose of meeting a specific need for a product representing a market opportunity. During the COVID‐19 pandemic ad hoc supply chains were built to meet the need for PPE, which provided companies with the opportunity of securing their business in times of crisis by retaining employees, generating profit, and making a contribution to society. Second, ad hoc supply chains are built in short order in response to an immediate need. Given the immediate need for PPE, companies did not plan these supply chains over a long time period as they would do in their core business. Instead, companies responded right away, even though they had to build the supply chains from scratch. Third, ad hoc supply chains are built for a limited period in accordance with the time‐limited need for PPE. For most of our sample companies, building ad hoc supply chains for PPE during the pandemic was an intermediate activity until the market and their core businesses recovered. After some months, by which time many customers were well supplied with PPE, they terminated the PPE business and refocused on their core business.

However, a few companies in our sample (CO21, CO23, CO24, CO26, and CO28) did convert their intermediate PPE into a new, long‐term PPE business. Although these companies integrated PPE production into their core businesses, ad hoc supply chains were only built for the limited time period during which they experienced time pressure. As soon as the time pressure fell away, these supply chains could no longer be viewed as ad hoc, since there was no longer a predominant immediate and specific need. Thus, we argue that ad hoc supply chains are operated up to a certain point, at which companies need to decide whether to stop production or integrate the supply chains into the core business. Either way, that is the point where the ad hoc character of the supply chains ends.

5.2. Supply chain agility

In our research context, supply chain agility is defined as the ability of a firm and its partners to respond rapidly to an uncertain supply chain disruption (Braunscheidel & Suresh, 2009; Lee, 2004; Swafford et al., 2006). The underlying logic here is that supply chain disruptions are highly uncertain to happen but have a high negative impact on the supply chain performance. Therefore, when facing a supply chain disruption, companies are required to act in a short time to reduce potential negative effects. The COVID‐19 pandemic disrupted global PPE supply chains. Due to the limited transportation capacity from China to Europe, export restrictions in many countries, and a lack of PPE manufacturing capacity in Europe the increased demand for PPE could not be met. This imbalance between supply and demand resulted in a global PPE shortage exposing many people at health risk. The demand for large volumes of PPE was addressed by manufacturing companies that ramped‐up PPE production so that the global PPE production output increased. While building ad hoc supply chains, companies were exposed to time pressure caused by the urgent need for PPE. Worldwide, large quantities of PPE were required immediately to curb the spread of COVID‐19. In hospitals and other healthcare facilities in particular, PPE was essential to protect frontline workers against infection. But also, individuals needed masks, especially those with a pre‐existing condition and elderly people.

In response, companies demonstrated extreme speed when building ad hoc supply chains for PPE: New partners were found within hours, products were developed overnight, new materials were sourced within days, and production machinery was engineered within weeks. Indeed, all our sample companies built these ad hoc supply chains faster than they would have built new supply chains in their core business. While all our sample companies were extremely quick, they differed in the extent to which they realized supply chain agility. Nevertheless, we did not measure the different levels of speed at which the companies built ad hoc supply chains. We regard speed to be binary in the sense that we only analyze companies that built ad hoc supply chains at the beginning of the COVID‐19 pandemic and thus responded extremely fast to the demand shock for PPE. Therefore, we conclude that all sampled companies realized supply chain agility when building ad hoc supply chains.

5.3. Dynamic capabilities

Our data suggest that the companies building ad hoc supply chains rely heavily on their dynamic capabilities. These dynamic capabilities were used to create a new configuration of operational capabilities required for building ad hoc supply chains. An operational capability, also termed “zero‐level” capability (Winter, 2003, p. 992), is a company's ability to deploy its resources with the goal of generating sales in the core business by performing highly routinized operational processes (Amit & Schoemaker, 1993; Collis, 1994; Helfat & Peteraf, 2003; Helfat & Winter, 2011; Winter, 2003). Dynamic capabilities, instead, build, integrate, and reconfigure these operational capabilities (Helfat & Peteraf, 2003; Teece et al., 1997). This results in new capability configurations (Eisenhardt & Martin, 2000) by selecting and combining the most effective operational capabilities with which valuable new tasks in changing environments can be performed (Eisenhardt & Martin, 2000; Zahra et al., 2006). In this sense, dynamic in the term “dynamic capabilities” means that operational capabilities are reconfigured based on what the company requires to address a change (Amit & Schoemaker, 1993; Collis, 1994; Eisenhardt & Martin, 2000; Helfat & Peteraf, 2003; Helfat & Winter, 2011; Teece et al., 1997; Winter, 2003). This is why dynamic capabilities are also termed “higher order” capabilities (Winter, 2003, p. 992). In our data, we observe that companies deployed dynamic capabilities to purposefully select internal and external capabilities developed in their core businesses that enable them to build ad hoc supply chains.

5.3.1. Internal capabilities

We found leveraging knowledge allowed companies to create and operate ad hoc supply chains. Existing knowledge developed in their core businesses was a major resource that gave companies an advantage over other companies that first had to acquire this knowledge (Grant, 1996; Kogut & Zander, 1992). In particular, procurement knowledge was useful when buying raw materials and machines for PPE production. For instance, sourcing raw material in a short market was facilitated by the companies' expertise with textiles, meltblown materials or chemicals. Furthermore, their cultural experience enabled many companies to accelerate their sourcing activities in Asia. Developing the product and finding suppliers was facilitated and accelerated by product knowledge. Chemical companies, for example, could rely on knowledge about ethanol, which is required both in their core business and in the production of disinfectants. The production ramp‐up phase was accelerated by manufacturing knowledge. For example, chemical companies started disinfectant production as they were familiar with processing chemical ingredients and had prior contact with relevant suppliers. Similarly, companies in the automotive, textile, and fashion industries that ramped‐up production of face masks were accustomed to cutting and processing textiles in their core business.

Similarly to leveraging knowledge, all companies could rely on initial capacity, defined as the maximum output a system can realize; for example, manufacturing capacity is a measure of how much a production system can produce (Olhager et al., 2001; Plambeck & Taylor, 2005). Exploiting capacities was evident in companies that utilized internally available machines, equipment, and material. Given that all the sample companies operated in core businesses, the required capacities they subsequently used for building ad hoc supply chains were (in part) already available. For example, manufacturers repurposed existing machinery and equipment, such as the welding or sewing machines used to produce automotive textiles or fashion items. Furthermore, companies could use materials already in stock in their warehouses and access spare workforce caused by the shutdown of the core business. All available production capacity was used to its maximum extent to increase the production utilization rate, often operating nonstop day and night and at weekends. Moreover, companies extended their production capacity, for example, by employing more workers or buying additional machinery and equipment.

Engaged employees (Kahn, 1990; Schaufeli et al., 2002) who were committed (Meyer & Allen, 1991) and motivated (Ryan & Deci, 2000) also enabled companies to build ad hoc supply chains. Committed employees showed a high level of support for new and speedy market entry, based on a common understanding that PPE would help curb the COVID‐19 pandemic. They were proud to be contributing to such highly important project work. Although the new project meant significant changes to their routine work, individuals accepted the new working conditions. Moreover, by pushing their limits and doing additional work, such as helping coworkers and voluntarily working overtime and at weekends, they expressed a high willingness to perform. In addition to work commitment, the employees in our case companies developed the motivation to become engaged in work activities related to building ad hoc supply chains. Extrinsic motivation was demonstrated by many employees who engaged in project work out of gratitude for their jobs, given the hazards of short‐term work or the risk of unemployment during the crisis caused by the pandemic. Intrinsically motivated employees enjoyed the project work and wanted to help fighting the pandemic.

While many companies could already rely on engaged employees, in some cases leadership skills were required to ensure employees engaged in the new project (Kets de Vries, 1994). When the PPE initiative began, many employees were dissatisfied with the new work situation. Lacking the flexibility to change routines and depart from standard processes, many employees opposed the project or even hated it. In addition to these mental barriers, many employees did not believe in their firms' capabilities. Dissatisfaction among employees festered due to evolving conflicts within the (cross‐functional) teams and the increasing mental strain imposed by new work tasks and the challenges of performing under high time pressure. To engage employees in work, some managers displayed emotional intelligence (Mayer & Salovey, 1993). Project leaders addressed the emotions of team members by expressing care for employees and demonstrating empathy. For example, some managers had to evince a great deal of caring to alleviate the concerns of (individual) employees and convince them of the viability of the change. Empathy was useful when introducing changes in routines such as a new shift model, including night work. In addition to emotional intelligence, some leaders revealed team‐building skills by purposefully selecting team members for the work on ad hoc supply chains to prevent conflict. For example, in some companies, troublemakers who could block the project were excluded from work.

5.3.2. External capabilities

Companies activated their networks to access capabilities they lacked internally. Without the time needed to develop such capabilities themselves, the firms searched their networks for partner companies with complementary assets and knowledge (Gulati et al., 2000; Lorenzoni & Lipparini, 1999). In their business networks (Boso et al., 2013), companies looked for both regular and new business partners often on online platforms developed by business associations. Managers also used their social networks (Boso et al., 2013) including professionals, family and friends to look for PPE suppliers and customers. A significant driver of the cooperativeness revealed by network members was the high level of solidarity. Some network partners helped each other to source raw materials, manufacturing companies shared their product designs, and transportation companies offered help in distributing PPE.

With the challenge of building supply chains for products they had never produced before, trusted collaboration with external organizations allowed companies to compensate for capabilities not yet developed internally (Cao et al., 2010; Mesquita & Lazzarini, 2008). Both horizontal and vertical collaboration (Mesquita & Lazzarini, 2008) were used to create ad hoc supply chains. For example, vertical collaborations were established between textile companies that had access to raw materials and contract manufacturers with production capacities. Other companies produced PPE in collaboration with healthcare companies that distributed PPE products through their channels. Horizontal collaborations were established to share information or resources for product design and sales. For instance, healthcare experts and research facilities helped manufacturing companies design products, and consulting companies assisted in developing the market. Such collaboration was reinforced by a high level of trust between participating parties (Fawcett et al., 2012).

To mitigate the prevailing uncertainties during the pandemic, some companies aimed for supply chain reliability (Li et al., 2008; Lukinskiy et al., 2014). A reliable supply chain helped companies realize production by reducing the risks of a lack of raw materials, unavailable production machinery, and the inability to deliver. Supply reliability enabled some companies to create stock as a buffer against future supply shortages. Using multiple suppliers and contractually securing supplies also increased the reliability of their supply of raw materials and production machinery. For example, one company ordered machines from two different suppliers to ensure that at least one machine was available at the planned time. Sales reliability mitigated the high uncertainty caused by limited commitments and long‐term contacts. A few companies entered one‐off sales agreements, for example, agreements with governmental institutions. Others aimed to reach a broad market for sales by distributing to a wide variety of customers, including individual consumers, retailers, and federal authorities, through a variety of sales channels, such as newspapers, broadcasts, websites, and social media.

5.4. Entrepreneurial orientation

Our data indicate that companies building ad hoc supply chains demonstrated an entrepreneurial orientation, that is, an organizational direction that captures firm behavior and results in seizing market opportunities (Anderson et al., 2014; Covin & Slevin, 1991; Dess & Lumpkin, 2005; Lumpkin & Dess, 1996; Miller, 1983; Naman & Slevin, 1993). Entrepreneurial orientation is a multidimensional construct that is characterized by five indicative behaviors: Proactiveness, risk‐taking, innovativeness (Covin & Slevin, 1988; Lumpkin & Dess, 1996; Miles & Arnold, 1991; Morris & Paul, 1987), autonomy, and competitive aggressiveness (Lumpkin & Dess, 1996). In our study, we observed that our sample companies began proactively to build ad hoc supply chains early in the pandemic and invested heavily in capabilities and resources despite the numerous risks they faced in an uncertain market. Their competitive aggressiveness enabled them to rush into the market with fit‐for‐purpose products that were improved as the market requirements increased.

Proactiveness enabled these companies to sense the opportunity presented by the PPE demand shock and start PPE supply chains ahead of their competitors (Dess & Lumpkin, 2005; Lumpkin & Dess, 1996). Many companies anticipated the market need for PPE based on market observations and insights they gained from their networks, or simply on their own intuition. The awareness that large quantities of PPE would soon be needed prompted many companies to take action by adopting a “just do it” mentality and deciding to start PPE production before the actual need and incentives were articulated. Some companies did not even wait for their products to be certificated before producing them on a large scale and putting them on the market. Moreover, to start PPE production early, companies established contact with suppliers ahead of their competitors to secure raw materials. This proactive behavior was determined by feasibility confidence. Many companies reported that the decision to start PPE production resulted from their confidence in their existing capabilities.

Risk‐taking behavior enabled our companies to value PPE production as an opportunity, (Forlani & Mullins, 2000; March & Shapira, 1987) prompting them to invest in ad hoc supply chains despite the many uncertainties involved, for example, the non‐predictability of demand (Dess & Lumpkin, 2005; Lumpkin & Dess, 1996). In view of the dynamic PPE market, building ad hoc supply chains was a risky venture. Entering a new market for critical products created a risk for their core businesses, from which their focus was removed. Additionally, investing large amounts of money in machinery, equipment, raw materials, and workers in supply chains that were not guaranteed to pay off required a willingness to take risks. To speed up operations, some companies did not even sign contracts with customers and partners, trusting each other's word. Furthermore, companies took high risks when confidently starting production before obtaining official permission to do so, or selling products in advance of actual production.

Companies building ad hoc supply chains demonstrated competitive aggressiveness to enter the market quickly (Dess & Lumpkin, 2005; Lumpkin & Dess, 1996). In contrast to the market for their core business, the PPE market demanded rapidly available products. Therefore, the companies' initial focus was merely on pushing high volumes of fit‐for‐purpose products onto the market to exploit the initial demand. These products did not have to meet strict requirements to be sold and many companies, for example, avoided time‐consuming product certification or using high‐quality materials. However, as an increasing number of providers entered the PPE market and the first shipments from China arrived, consumers' requirements for PPE increased. To remain competitive, firms adapted their products to current market needs by improving the quality (e.g., by certifying the products or using high‐quality materials) and product designs (e.g., by individualizing face masks or changing their cut). Competitive aggressiveness was enhanced by efforts to understand the market. For example, companies watched competitors and suppliers in the PPE market to gain an understanding of trends, developments, and availability of material and machinery.

5.5. Temporary orientation

One construct, previously unknown in the literature, emerged from our data analysis, that of temporary orientation. We define temporary orientation as an organizational direction that captures firm behavior resulting in the exploitation of short‐term opportunities. Specifically, companies focus on speed to exploit the period of time in which ad hoc supply chains are needed. We identified four practices that companies implemented temporarily to speed up building ad hoc supply chains: Pragmatic problem‐solving, process alignment, geographical proximity, and assembling project teams.

Managers showed a pragmatic approach to problem‐solving, which allowed them to create rapid solutions tailored to their ad hoc supply chains. Pragmatic problem‐solving significantly reduces the complexity of problems and allows managers to find simple and practical alternative solutions (Bedell‐Avers et al., 2008; Mumford et al., 2008). For example, we observed simple solutions for developing the new products: Product designs were often kept as straightforward as possible to avoid multiple test cycles, enable certification, and facilitate production. Pragmatism was further manifested by finding alternatives for transportation, production machinery, suppliers, and raw materials. For example, one company delivered disinfectants via its inhouse chauffeur service, which was usually reserved for top management.

Processes are essential to increase efficiency through standardization; however, the ad hoc situation during the COVID‐19 pandemic did not fit with companies' time‐consuming standard processes. Process alignment allowed companies building ad hoc supply chains to avoid these standard processes. For instance, many planning processes were bypassed, an approach similar to what is known in theory as low process formalization (Buganza & Verganti, 2006; Taylor et al., 2014). Instead of following planning processes, companies often relied on intuition and improvisation when building ad hoc supply chains. For example, to reduce administrative effort and minimize process complexity, some companies worked with spreadsheets rather than ERP systems to plan demand and supply. Even if they did not completely bypass processes, many companies parallelized them, also known as concurrent engineering (Duffy & Salvendy, 1999; Swink, 1998). For example, companies designed products in parallel with ramping up production, or acquired customers in parallel with sourcing raw materials.

The PPE supply chain that emerged in Germany during the COVID‐19 pandemic was highly regional, with production companies, partners, and customers within close physical distance. Geographical proximity (Gray et al., 2015; Narasimhan & Nair, 2005) to suppliers reduced lead times, and cooperation with partners was accelerated by working on‐site. This was a significant time advantage, compared to sourcing from Asia, as was usually the case. Local supply chains, including manufacturing companies, suppliers, and end customers in Germany (and nearby in Europe), facilitated fast sourcing of raw materials and rapid distribution of finished goods to customers. In addition to accelerating supply and delivery, short distances between supply chain partners enabled close collaboration. For instance, when operating their ad hoc supply chain, managers from different companies worked together at one location, sometimes in dedicated project rooms.

Companies assembled project teams that worked jointly on the same tasks (Crawford & Lepine, 2013). Within these teams, staff often worked exclusively on project activities. Project teams were characterized by a centralized structure (Crawford & Lepine, 2013), small team size, and a flat hierarchy. However, there was often one dedicated project leader who coordinated teamwork. Although only a few employees were involved, skilled people from different functional units were included. To enable enhanced team communication, project teams had frequent, often daily, meetings to coordinate the ad hoc supply chains (Swink et al., 1996). Furthermore, team members communicated directly with each other by phone, text messages or video calls instead of by email. In addition to enhanced team communication, controlled autonomy allowed the team to make independent operational decisions while top managers set the project scope.

6. EMERGENT THEORETICAL MODEL ON AD HOC SUPPLY CHAINS

Our data suggest that companies facing business opportunities for products in high demand quickly build ad hoc supply chains with supply chain agility. We found that three theoretical constructs influence how supply chain agility affects the adoption of ad hoc supply chains: dynamic capabilities, an entrepreneurial orientation, and a temporary orientation. Figure 2 is a conceptual visualization of the theoretical model we developed. The fast adoption of ad hoc supply chains is realized by a company's supply chain agility. Dynamic capabilities enable companies to reach supply chain agility by reconfiguring internal and external capabilities in response to what is required to build ad hoc supply chains. Two strategic orientations guide companies to align their activities with the time pressure in the external environment: an entrepreneurial orientation allow companies to exploit the emerging opportunity of ad hoc supply chains by deploying the internal and external capabilities immediately and a temporary orientation allow companies to exploit the short‐term opportunity of ad hoc supply chains by adopting accelerating processes and structures for a limited time period. In this section, we discuss the interactions of these constructs.

FIGURE 2.

Theoretical model on ad hoc supply chains

6.1. Dynamic capabilities

Companies building ad hoc supply chains deploy specific internal and external capabilities that do already exist using dynamic capabilities. The purposeful selection of internal and external capabilities allows them to create a capability set from the existing internal and external capabilities that is relevant and suitable for building ad hoc supply chains. We found in our study that dynamic capabilities increase a company's supply chain agility as dynamic capabilities were essential to deploy the necessary internal and external capabilities the company developed in their core businesses. The fast deployment of these capabilities was only possible with dynamic capabilities. Given the immediate need for PPE, companies did not have sufficient time to develop new operational capabilities. They rather relied on existing capabilities from their core businesses which were reconfigured by dynamic capabilities and thus purposefully applied in a short time to build ad hoc supply chains.

Conversely to previous researchers (Altay et al., 2018; Chiang et al., 2012; Côrte‐Real et al., 2017; Felipe et al., 2016; Nijssen & Paauwe, 2012), we claim that supply chain agility is not a dynamic capability. We provide empirical evidence for the conceptual distinction between these constructs. As defined in this study, supply chain agility enables a firm to respond quickly to an environmental change, whereas dynamic capabilities allow companies to change by reconfiguring operational capabilities. However, dynamic capabilities and supply chain agility are undoubtedly strongly connected, as some researchers suggest (Blome et al., 2013; Teece et al., 2016). Furthermore, many researchers suggest that dynamic capabilities have a positive effect on a firm's speed and responsiveness (Teece, 2012; Teece & Pisano, 1994). Thus, we propose:

Proposition 1a

Dynamic capabilities reconfigure and deploy internal capabilities, enabling a company to realize supply chain agility when building ad hoc supply chains.

Proposition 1b

Dynamic capabilities reconfigure and deploy external capabilities, enabling a company to realize supply chain agility when building ad hoc supply chains.

6.2. Entrepreneurial orientation

Entrepreneurial orientation is a strategic orientation (Li et al., 2010; Wiklund & Shepherd, 2003; Zhou et al., 2005), “that direct and influence the activities of a firm and generate the behaviors intended to ensure its viability and performance” (Hakala, 2011, p. 199). Companies with a strategic orientation aim to align their activities in accordance with the external environment (Narver & Slater, 1990) to realize superior performance (Gatignon & Xuereb, 1997). Thus, strategic orientations act like a channel that captures the dynamics in the external environment and use this information to adjust the company's actions. In particular, in a dynamic environment that provides abundant opportunities, companies benefit from an entrepreneurial orientation (Miles & Arnold, 1991; Rosenbusch et al., 2013). To exploit these opportunities, entrepreneurial companies “commit resources” (Hakala, 2011, p. 202) in an effort to obtain a competitive advantage (Atuahene‐Gima & Ko, 2001). Since an entrepreneurial orientation directs a company to invest its resources and capabilities into the endeavor to exploit opportunities, we argue for a moderating effect of entrepreneurial orientation in the model of ad hoc supply chains. Specifically, we propose that in adopting ad hoc supply chains, an entrepreneurial orientation moderates the effect between a company's dynamic capabilities and its supply chain agility. During the COVID‐19 pandemic, a company that had an entrepreneurial orientation could deploy dynamic capabilities immediately. Their proactiveness enabled firms to enter collaborations and secure raw materials and machinery ahead of their competitors. Furthermore, risk‐taking behavior allowed companies to invest in resources, such as knowledge, money, assets, material, and their workforce, within a short time and under uncertainty to start PPE supply chains immediately. Competitive aggressiveness also allowed companies to reduce time‐to‐market by developing fit‐for‐purpose PPE that cut development time significantly. We conclude that these entrepreneurial behaviors allow companies to speed up the use of dynamic capabilities in response to the immediate need for ad hoc supply chains. Thus, we make the following proposition:

Proposition 2

An entrepreneurial orientation positively moderates the effect of dynamic capabilities on supply chain agility when building ad hoc supply chains.

6.3. Temporary orientation

Analogous to entrepreneurial orientation, we view temporary orientation as a strategic orientation that moderates the effect of supply chain agility on ad hoc supply chains. A temporary orientation allows a company to adapt behaviors that accelerate building ad hoc supply chains according to a time‐limited need (for PPE in our case). In anticipation of saturation of the PPE market in the near future, the companies in our study pursued an opportunity to build ad hoc supply chains. To exploit the limited time frame in which ad hoc supply chains could deliver added value, companies with a temporary orientation were able to align with the need for speed by adopting temporary behaviors to build ad hoc supply chains within the fastest time possible. Specifically, pragmatic problem‐solving speeded up decision‐making; process alignment enabled companies to avoid time‐consuming standard processes; suppliers and partners within close physical distance facilitated collaboration; and project teams enabled a time‐efficient focus on building ad hoc supply chains. A temporary orientation allows a company to adapt its organizational structures and processes for a limited period of time until specific goals are realized. Consequently, we propose the following:

Proposition 3

A temporary orientation positively moderates the effect of supply chain agility on the adoption of ad hoc supply chains.

7. DISCUSSION

7.1. Emergent theoretical model discussion

The purpose of this study was to explain how companies realize supply chain agility when building ad hoc supply chains under time pressure. Based on empirical data collected in 34 German companies during the COVID‐19 pandemic, we developed an emerging theory that identifies dynamic capabilities, entrepreneurial orientation, and temporary orientation as positive factors that enable companies to leverage supply chain agility. This theory makes two major contributions to theory on ad hoc supply chains and several contributions to different literature streams.

First, we advance the understanding of ad hoc supply chains, an emerging phenomenon that is often observed in times of crisis, such as in humanitarian disasters, war, or, as in our study, pandemics. Ad hoc supply chains have a very different character than regular supply chains operated in core businesses since they are built to answer specific, immediate, and limited needs. Our research provides an initial theoretical understanding of these emerging supply chains, which are of great importance for mitigating the impact of disrupted supply chains.

Second, we present a theoretical model explaining how companies build ad hoc supply chains by using supply chain agility. The theoretical model identifies three interactive capabilities that enable companies to accelerate the process: Dynamic capabilities, an entrepreneurial orientation, and a temporary orientation. In particular, dynamic capabilities allow companies to build supply chains for a specific need by configuring a capability set needed to perform these activities. This view is underpinned by Helfat et al. (2007), who stress that dynamic capabilities, in contrast to operational capabilities, “have some implicit aim even if not fully planned” (Helfat et al., 2007, p. 5). An entrepreneurial orientation allows a company to start building ad hoc supply chains immediately to meet an immediate need. The entrepreneurship literature considers that an immediate response allows companies to generate first‐mover advantages that are hard to imitate and so generate a high rent from the opportunity (Rumelt, 2005, pp. 19–22). Finally, a temporary orientation allows companies to adopt a behavior for a limited period of time, so that operations are accelerated. The temporary structure is then dropped as soon as goals are reached. This view aligns with the project management literature describing a project as an organizational endeavor that is of “temporary existence, being disbanded when the new state is achieved” (Turner, 2009, p. 3).

Furthermore, our study offers several insights within the scope of dynamic capabilities theory. In our theoretical model, we identified dynamic capabilities that enable companies to reconfigure and deploy internal and external operational capabilities, both of which are needed to build ad hoc supply chains. In doing so, we provide empirical evidence for two much‐discussed characteristics of the dynamic capabilities construct. First, there is an ongoing debate about whether or not dynamic capabilities are primarily routinized (Wenzel et al., 2021). There seems to be widespread consensus in a large part of the literature that dynamic capabilities are stable and routinized through their repeated application in practice (Eisenhardt & Martin, 2000; Helfat et al., 2007; Helfat & Peteraf, 2003; Helfat & Winter, 2011; Teece, 2007; Teece et al., 1997; Winter, 2003; Zollo & Winter, 2002; Zott, 2003). However, our data suggest that dynamic capabilities cannot be routines when deployed in highly volatile environments like the COVID‐19 pandemic. We argue that the companies that built ad hoc supply chains to provide PPE during the pandemic could not rely on any previously developed supply chain building routines. Our observation supports the different perspectives of some researchers on the process‐character of dynamic capabilities, highlighting the improvised and, unstable characteristics of dynamic capabilities (Eisenhardt & Martin, 2000; Schreyögg & Klisch‐Eberl, 2007; Teece, 2012; Teece et al., 2016). Moreover, we argue that uncertainty in dynamic environments actually precludes the usefulness of routines. In a rapidly changing environment, such as the companies in our study experienced during the COVID‐19 pandemic, routines do not help responses to frequent and unanticipated changes. Consequently, we claim that dynamic capabilities do not derive from routines when applied in dynamic environments.

Second, we found that for ad hoc problem‐solving, dynamic capabilities are required to start a new business at short notice. This contradicts the argument made by Winter (2003) that ad hoc problem‐solving and dynamic capabilities are distinct approaches that are not mutually compatible. Our data shows that ad hoc problem‐solving and dynamic capabilities are compatible and moreover that dynamic capabilities were essential for companies to build ad hoc supply chains for PPE. Our work is in line with prior studies that suggest an interdependence between ad hoc problem‐solving and dynamic capabilities. Teece (2012, p. 3), for example, denotes Winter's (2003) assumption that these approaches contradict each other as a “false dichotomy” and argues that dynamic capabilities have an ad hoc element. Furthermore, Ritala et al. (2016, p. 9) suggest “that a total separation of dynamic capabilities and ad hoc approaches may not be warranted in unfamiliar problem situations faced by firms and that the two approaches may be complementary and parallel.”

Next, we contribute to the entrepreneurship literature by identifying entrepreneurial orientation as a moderator between dynamic capabilities and supply chain agility. Although acknowledged in the literature to be a dimension of the entrepreneurial orientation construct (e.g., Dai et al., 2013; Lee et al., 2001), innovativeness—in the sense of new products, services, technology, and/or processes to the market (Garcia & Calantone, 2002)—did not emerge in our data. Starting production of PPE already known to the market was not an innovation. Moreover, innovations could be counterproductive; for example, one company developed a rubber mask with a changeable filter to differentiate its product even though the market demanded only standard PPE. Our insight is supported by Schneider and Hall (2011), who argue that the absence of a market for an innovative product is one reason why market entries fail. We therefore conclude that innovativeness is neither effective nor required for building ad hoc supply chains.

Similarly, we did not identify autonomy as a dimension of entrepreneurial orientation in our data but as a dimension of temporary orientation, demonstrated in the capability to assemble project teams. In the original sense in which Lumpkin et al. (2009) use it, autonomy in entrepreneurial orientation is understood to be strategic autonomy, that is, the extent to which a team makes independent decisions about a problem. Lumpkin et al. (2009) distinguish strategic autonomy from structural autonomy, which is about work procedures and team structure and does not characterize a company's entrepreneurial orientation. Our data provide evidence for structural autonomy temporarily established by managers for project teams working on ad hoc supply chain activities. Project team members were allowed to make independent decisions only within the project scope defined by top‐level managers. As such, autonomy was identified as a characteristic of temporarily assembled project teams.

Finally, we developed the construct of temporary orientation, which moderates the relation between supply chain agility and the adoption of ad hoc supply chains. We distinguish temporary orientation from the short‐termism known in strategic decision‐making research. Short‐termism is a firm's “course of action that is best for the short term but suboptimal over the long run” (Laverty, 1996, p. 826). Short‐termism grasps the interplay between a firm's behavior with respect to promising outcomes in the near future but with “detrimental consequences for the long term” (Marginson & Mcaulay, 2008, p. 274). In this sense, short‐termism describes the tendency of firms to focus on the short‐term benefits of decisions without assessing their (potentially negative) long‐term consequences. We argue that short‐termism does not capture the value of a temporary orientation that we have observed in companies building ad hoc supply chains. Short‐termism has negative connotations assuming that a focus on short‐term advantages has a detrimental effect on the long‐term future of the company. Our data do not indicate that building ad hoc supply chains has a negative effect on the future business. In the context of the COVID‐19 pandemic, the companies considered PPE supply chains an opportunity when they were experiencing a shutdown of their core businesses and a huge demand for PPE. In the long run, the temporary businesses that operated for a limited time period did not negatively affect the core business, and even provided the opportunity to enter a new business field.

7.2. Cross‐case discussion

Following the discussion of the theoretical model that was grounded in our data, we now discuss some additional observations across the cases. Since our exploratory data do not allow for a formal cross‐case analysis, these additional observations may inspire further research in this emerging area of research. Online Appendix IV includes a table with a complete set of relevant observations. We highlight three of these in the following discussion.

First, despite the initially limited need for ad hoc supply chains, certain companies used them to develop a new long‐term business for two potential reasons. On the one hand, ad hoc supply chains built for unknown products allow companies to expand the company's product portfolio. Some of our data suggest that for companies operating in a struggling industry, for example, the automotive industry, PPE production was a great opportunity to diversify their core business. On the other hand, we observed that some companies that invested heavily in mask production machinery tended to continue production in their core business. Therefore, the capital invested into ad hoc supply chains might influence a company's decision whether to continue ad hoc supply chains or not. Thus, despite the time‐limited need for ad hoc supply chains, they might provide the opportunity for companies to develop a new long‐term business.

Second, even though all companies realized supply chain agility, our data suggest that some companies were faster than others. We suspect that the agility companies have on hand from their core business determines the level of agility realized when building ad hoc supply chains. Some companies implied that they are accustomed to rapidly changing environments, which helped them to speed up building ad hoc supply chains. For example, an interviewee from an automotive supplier said that rapid response to customers' changing demands was an everyday part of their business. Thus, prior development of supply chain agility might affect a company's capability to build ad hoc supply chains at short notice.

Third, we observe that government decisions might influence the competitive situation in the markets. As reflected by our company sampling, most of the German companies built ad hoc supply chains for face masks and disinfectants although gowns were also short in supply. As a result, companies building ad hoc supply chains for masks and disinfectants faced strong competition in the market, while those that began producing gowns were out of competition. We suppose that incentives provided by the government determined the market competition since the German government did not provide incentives for gown production but instead offered financial and regulatory support for mask and disinfectant production. This lack of incentives is a plausible explanation why fewer companies started gown production, which then resulted in a low competition in the market.

8. CONCLUSION

In the early stage of the COVID‐19 pandemic, companies built ad hoc supply chains for PPE at an extreme speed. The time pressure caused by the unusual demand shock for potentially life‐saving PPE prompted companies to start new businesses in an unfamiliar market by developing new products, contracting new suppliers, and distributing to new customers within a short time. In this paper, we have leveraged this unique situation and our access to the leadership of companies that conducted this massive operation by developing an understanding of how companies realize supply chain agility when building ad hoc supply chains under time pressure. We provide an initial understanding of how dynamic capabilities, entrepreneurial orientation and temporary orientation affect supply chain agility in the creation of ad hoc supply chains. Interestingly, while our setting is unique, the data provide evidence to help advance ongoing discussions in the theories relating to dynamic capabilities. In particular, we show that approaches based on dynamic capabilities and approaches based on ad hoc problem‐solving are not mutually exclusive. We argue that dynamic capabilities actually enable ad hoc problem‐solving approaches, especially under time pressure. This lends supports to the argument that dynamic capabilities are not necessarily routine processes. In addition, entrepreneurial orientation has been shown to moderate the effect of dynamic capabilities on supply chain agility. Companies with an entrepreneurial orientation can deploy dynamic capabilities immediately. In particular, proactive, risk‐taking, and competitive aggressive behavior enable a firm to deploy internal and external capabilities in the shortest time but not innovativeness and strategic autonomy. Further, our study has allowed us to develop the construct of temporary orientation. We show that companies that build an ad hoc supply chain that they know will be time‐limited are able to act faster. This new insight may help develop structures and processes for companies that operate in markets with rapidly changing external conditions. Much of the emphasis in the literature is on the ability to respond rapidly (supply chain agility), but we argue that speed can also be enhanced by having a finite time horizon.

Regarding the context of our data collection, it is important to note that the incentives provided by the German government were a significant driver for building ad hoc supply chains. Financial support and relaxed product and manufacturing regulations encouraged companies to contribute to PPE production. Furthermore, the lack of a long‐term political strategy for local PPE production in Germany impeded the implementation of a local PPE supply chain infrastructure that would be needed to reduce the risk of future global supply chain disruptions. Based on these insights, we assume governments have a critical role in enabling companies to build ad hoc supply chains if barriers, like essential investments and restrictive regulations prevent market entry.

Our research should be regarded as an initial promising step toward an understanding ad hoc supply chains. We suggest three relevant avenues of future research to develop this emerging theory further. First, we argue that defining the time frame of a temporary orientation is critical to realizing speed in building ad hoc supply chains. Realizing supply chain agility is a very cost‐ and effort‐intensive venture. As soon as initial market saturation is reached, it makes sense for companies to return to more efficient and sustainable operations. For companies, the critical decision is then whether to proceed with the new business and, likewise, when to transition ad hoc supply chains into a regular business. Second, we assume that the existing degree of supply chain agility in companies influenced the speed with which they built ad hoc supply chains, since we were unable to observe sufficient details about the actual level of speed realized. Further research focusing on the speed of ad hoc supply chains in relation to previously developed supply chain agility could help the evaluation of ad hoc supply chains. Third, we propose to explore more performance indicators of ad hoc supply chains. While we focused on agility as an enabler of ad hoc supply chains, we assume that common supply chain goals, for example, profitability and quality, are not completely deferred. Studying the trade‐off between speed and strategic goals in building ad hoc supply chains under time pressure might be promising to further examine the success of ad hoc supply chains.

Supporting information

Appendix S1 Supplementary Information

APPENDIX A.

A.1. INTERVIEW GUIDELINES—FIRST INTERVIEW ROUND

Part 1: Characteristics of ad hoc supply chains

What PPE is your company producing during the COVID‐19 pandemic? Why did your company start PPE production? What skills or expertise do you particularly benefit from when producing PPE? What do you need for production? E.g.: Partners/suppliers Machinery Materials Certification What resources were already in place before the crisis and what has now been sourced at short notice? Who designed the product? How do you decide how much PPE to produce? What tool do you use for production planning? Who are your customers and how did you find them? How do you distribute the products to your customers?

Part 2: Challenges of ad hoc supply chains

11 What challenges did you face in establishing the new supply chain? E.g.: Identification of suppliers/partners Availability of components / ingredients Competition in procurement of materials Certification of products Product quality Acquiring new skills and know‐how Financial resources for investments Production bottlenecks Access to customers 12How did you overcome these challenges? 13What problems have you not been able to solve so far?

Part 3: Outlook and learnings

14How long will you continue producing PPE? 15What did you learn from building the new supply chains? 16What positive conclusions do you draw from building PPE supply chains?

A.2. INTERVIEW GUIDELINE—SECOND INTERVIEW ROUND

Part 1: Status quo of PPE production

What has happened since our last interview? Are you still producing PPE? If production is continuing: How long will you keep producing PPE?/If production has already stopped: Why did you stop producing PPE? ➔ Follow‐up questions depending on the interview partner

Part 2: Reflecting on the speed of building ad hoc supply chains

4Compared to your core business, did you build supply chains particularly quickly during the crisis? 5What is the usual lead time between product development and beginning production? 6How do you think your company was able to realize this speed?

Part 3: Outlook and learnings

7What lessons did you learn from realizing this speed that could be applied to your core business? 8Are there any situations in your core business where you will need this speed? 9What incentives would the government have to give you to continue production? 10In retrospect, would you enter the PPE business again? If so, why? If not, why not?

A.3. INTERVIEW GUIDELINES FOR EXPERTS—FIRST INTERVIEW ROUND

Part 1: Status quo of healthcare supply chains

What healthcare products does your company produce? How did production/demand change during COVID‐19 pandemic? Whom have you supplied during the COVID‐19 pandemic? What challenges does your company face?

Part 2: Reflecting on supply chain agility

5Compared to your core business, are you acting more rapidly during the pandemic? 6How does the government support companies that have started PPE production?

Part 3: Outlook and learnings

7What have you learned during the COVID‐19 pandemic? 8Do you think that local PPE production will be continued after the crisis?

A.4. INTERVIEW GUIDELINE FOR EXPERTS—SECOND INTERVIEW ROUND

Part 1: Status quo of healthcare supply chains

How did the PPE market develop? How did your business develop? How did PPE prices develop? How did the demand for PPE develop? Are the production incentives provided by the government still valid?

Part 2: Reflecting on supply chain agility

6Compared to your core business, did you act more rapidly? 7If yes, how did you demonstrate the increased speed?

Part 3: Outlook and learnings

8What did you learn during the COVID‐19 pandemic?

Müller, J. , Hoberg, K. , & Fransoo, J. C. (2022). Realizing supply chain agility under time pressure: Ad hoc supply chains during the COVID‐19 pandemic. Journal of Operations Management, 1–24. 10.1002/joom.1210