Procurement under pressure: shifting governance strategies in turbulent times

ABSTRACT

Organizations in the public sector generally use contractual governance mechanisms to govern buyer–supplier relationships. Little is known about what happens when procurement is put under pressure and which governance strategies are used in such turbulent times. News articles, parliamentary documents, and internal communication regarding the procurement of personal protective equipment during the COVID-19 crisis in the Netherlands were analysed to address this question. The findings show a brief shift from contractual to relational governance, followed by an attempt to return to contractual governance, albeit with different procedures. After the contract closes, limited governance strategies are applied, resulting in ‘laissez-faire’ governance.

KEYWORDS:

• Public procurement
• crisis
• governance strategies
• contractual governance
• relational governance
• healthcare procurement
• COVID-19

Introduction

Procurement is an important yet underestimated activity in the public sector. Governments, public organizations, and private organizations with public tasks (such as hospitals) procure goods and services for their own operation (e.g. laptops) and to execute their public tasks (e.g. hospital equipment, schoolbooks and highways). Consequently, procurement has large implications. Despite its clear societal and economic importance, procurement is still something few people notice until it goes wrong (McKee 2020).

In procurement, the relationship between buyer and seller requires governance. Two main governance strategies are distinguished: contractual and relational governance. Public procurement in Europe strongly leans towards the use of contractual governance (Grandia 2018). Contractual governance involves the use of formal contracts to govern buyer–supplier relationships (Akintoye, Beck, and Hardcastle 2008; Brown, Potoski, and Van Slyke 2016). For public organizations, the procurement process is guided by strict regulations and procedures laid down in European legislation. Detailed contracts form the basis of any buyer–supplier relationship (Bovis 2012; Grandia and Volker 2023). An important limitation of contracts is their inherent incomplete character (Brown, Potoski, and Van Slyke 2016). Consequently, contractual governance strategies might work fine under normal circumstances, but struggle with unexpected events due to their inflexibility and inability to cover all potential circumstances and developments (Davis 2007). Relational governance offers an alternative, based on the notion that inter-organizational exchanges are embedded in social relationships and opportunistic behaviour can be prevented through information sharing and joint problem-solving (Granovetter 1985; Poppo and Zenger 2002). Unlike contractual governance, relational governance is considered better equipped to deal with uncertainty and complexity and can stimulate actors to adopt a more flexible attitude (Ibid.). Contractual and relational governance are not mutually exclusive. Mixing the two has been found to be an important key to success (Warsen, Klijn, and Koppenjan 2019).

Currently, turbulent times confront organizations with various crises, such as climate, healthcare, and financial crises (Ansell, Sørensen, and Torfing 2021). Crises can disrupt key aspects of patterns of social, organizational, and political interactions and influence how government operates and governs (Ansell et al. 2010; ‘T Hart, Rosenthal, and Kouzmin 1993). A crisis puts procurement processes under political, societal, time and economic pressure. Recent studies have emphasized the importance of public procurement in times of crisis and called for a new and robust strategic approach to managing buyer–supplier relationships (McEvoy and Ferri 2020; Phillips et al. 2022; Rosell and Saz-Carranza 2020). However, little attention has been paid to the governance strategies used to govern public procurement processes, and how these strategies react to the pressure of a crisis. Are governance strategies robust and flexible enough to safeguard the procurement process in times of crisis? With this paper, we aim to fill that gap. The research question therefore is as follows: ‘Which governance mechanisms are used in the public sector to govern the buyer–supplier relationship during a crisis?’

To answer this question, a recent crisis was selected where public procurement was at the forefront: the COVID-19 crisis (Sanchez-Graells 2020; Atkinson et al. 2020). This crisis is considered ‘a game changer’ for public administration, as it reveals the demand for robust governance strategies to deal with turbulent issues (Ansell, Sørensen, and Torfing 2021). The COVID-19 pandemic exerted immediate political, societal, and time pressures on procurement to safeguard healthcare professionals from illness and death (McEvoy and Ferri 2020). This study focuses on governance strategies during the procurement of personal protective equipment (PPE), such as surgical masks. Initially, a non-critical product, it became globally scarce at the crisis’s outset (Arrowsmith et al. 2021; Finkenstadt and Handfield 2021). We analyse PPE procurement in the Netherlands, where even this advanced market faced a severe PPE shortage (Tip et al. 2022), making it a relevant case of procurement under pressure. Via an extensive document analysis of news articles, parliamentary documents, and internal communication regarding the procurement of PPE, this inductive study aims to contribute to both governance and public procurement literature by providing insight into the use of governance mechanisms in managing buyer and supplier relationships in times of crisis.

In the remainder of this paper, we first present the sensitizing concepts that guide our thinking: procurement in a public sector context and contractual and relational governance. We then explain our methodological approach. The result section presents our core findings regarding the use of governance mechanisms and the shift therein. Finally, the conclusion and discussion provide an answer to the research question and a reflection on our findings.

Procurement in a public sector context

If organizations require a product or service, they can either produce it themselves or procure (buy) it. The procurement process typically has three phases: First, prepare, starting with the identification of the need for a good or service and the make-or-buy decision (explore and initiate) (Van Weele 2005). Second, purchase, including requesting and assessing offers (tender and assess), and the subsequent performing phase where a contract is drafted, signed, and executed (Grandia and Volker 2023). Finally, perform (implement, execute, and extend) as the supply or work needs to be managed and monitored until the end of its lifecycle (Patrucco, Luzzini, and Ronchi 2017). The procurement rules and regulations that apply vary significantly among country (public or private) sector and the ‘publicness’ of an organization (Telgen, Harland, and Knight 2012). Whether an organization is considered a contracting authority required to follow public procurement rules is determined by how they are funded, the level of political control and the extent to which they perform public and private tasks. Whether an organization, such as healthcare institutions, is considered a public contracting authority therefore varies per country (Grandia and Volker 2023). The Netherlands is known for its almost exclusively private, non-profit provision of care, e.g. via private non-profit hospitals or nursing homes (Bos, Kruse, and Jeurissen 2020; Jeurissen 2010). While these private healthcare institutions have a public task and are non-profit, they are not considered public contracting authorities in the Netherlands and thus do not have to abide by national and European public procurement regulations. In the case of COVID-19, public contracting authorities were allowed to deviate from regular public procurement rules as the EU Commission offered guidance on COVID-19 related procurement identifying it as an extreme and unforeseeable urgency for which the EU Directives did not contain procedural constraints (Folliot Lalliot and Yukins 2020; Telles 2022). Due to these developments, the boundaries between public and private sector organizations with respect to their procurement procedures are blurring. In this paper, we therefore examine the governance strategies employed in the procurement of PPE by actors in a public sector context. This encompasses procurement conducted by semi-public or private organizations with a public task, like private hospitals or nursing homes and by public contracting authorities, including ministries and academic hospitals.

Governance strategies in procurement

To prevent opportunistic behaviour and ensure a fair procurement process, the relationship between buyer and (potential) supplier must be governed. Two primary strategies are identified and used in public procurement for governing inter-organizational relationships: contractual and relational governance (Rosander 2022; Warsen 2021). Contractual governance, influenced by economic theories like transaction cost and principal–agent theory, assumes rational behaviour and the risk of opportunism (Akintoye, Beck, and Hardcastle 2008; Brown, Potoski, and Van Slyke 2016). Contracts aim to prevent opportunism, aligning the supplier’s behaviour with the buyer’s goals in procurement. Contracts include transaction rules (Brown, Potoski, and Van Slyke 2016) outlining both parties’ responsibilities, product requirements, monitoring, sanctions, risk, performance management, and force majeure or contract termination options (Warsen 2021).

Contrary to contractual governance, relational governance assumes that interdependencies between buyer and supplier prevent opportunistic behaviour from occurring. Relational governance can be considered ‘facilitative behaviours to foster the development of mutually beneficial relationships’ (Chuang et al. 2020, 258), such as voluntary information sharing, open communication, and joint problem-solving (e.g. Macneil 1980; Poppo and Zenger 2002). Trust plays a vital role in the perceived satisfaction with the relationship and performance. For example, trust grounded in a belief in a counterpart’s reliability has been found to increase satisfaction with the relationship and reduce opportunistic behaviour (Chuang et al. 2020). As relational governance can stimulate actors to adopt a more flexible attitude, it is considered a better fit for procurement in crisis situations than contractual governance. Crises are fundamentally distinct from ‘routine emergencies’ such as traffic accidents or fires and threaten core values of a social system, disrupting patterns of social, organizational and political interactions and influences how government operates and governs (Boin and Rhinard 2008; Rosenthal, Charles, and Hart 1989). Governance strategies underpinned by contracts may not be sufficiently flexible to govern relationships in the fast moving, rapidly changing environment of a crisis (Pilbeam, Alvarez, and Wilson 2012). It is important to note here that although contractual governance is considered the standard governance strategy in public procurement, public buyers have increasingly applied relational contracting to tackle growing demands of efficiency and improve the performance, in particular of large complex infrastructure projects (Rosander 2022). Moreover, although relational and governance strategies are based on different assumptions, they are not mutually exclusive. They can be mixed, and the mix can affect the performance of the project (Warsen, Klijn, and Koppenjan 2019). Mixing or changing governance strategies could thus result in a more successful procurement process. Studies examining responses during the COVID-19 crisis, for example, did find that more relational governance mechanisms were used by SMEs and businesses to cope with the emergent and sudden government regulations which affected business relationships (Ashiru et al. 2022; Obal and Gao 2020). However, little is known about the governance strategies in procurement in a public sector context during a crisis. In the next section, we therefore explain how we examined the governance of PPE procurement during the recent COVID-19 crisis.

Methodology

Public procurement during a pandemic

From procurement perspective pandemics and epidemics are the most critical types of crises, due to the direct impact of procurement on human lives (Harland et al. 2021). The COVID-19 crisis put unprecedented pressure on procurement to fulfil the demand for PPE and save lives. This led to significant political and media attention, allowing for content analysis of a large amount of news articles, and parliamentary documents. A case study was considered the best research strategy for this research, as it allows for an intensive study of a single unit for the purpose of understanding a larger class of units (Gerring 2004). In this research, we studied the procurement of PPE in the Dutch Healthcare sector during the COVID-19 crisis, to understand how governance mechanisms are used in the public sector to govern the buyer–supplier relationship during a crisis. We opted to conduct this research within the Dutch context, as the Netherlands has an advanced and well-established market that nonetheless experienced great challenges in the procurement of PPE during COVID-19. Moreover, in the Netherlands, public procurement is predominantly focused on the use of contractual governance mechanisms, but the Dutch consensus-based culture also allows for the use of relational governance mechanisms (see Warsen et al. 2020) making this a suitable context to study the use of both mechanisms in the procurement of PPE. Finally, the availability of documents regarding governance mechanisms to procure PPE in times of crisis was high in the Netherlands, as the Dutch Open Government Act regulates the right to actively disclose information to the public.

Data collection

Document analysis was selected as the most suitable research method due to its coverage, exactness, lack of obtrusiveness and reactivity (Bowen 2009; Karppinen and Moe 2019; Yin 1994). Even though document analysis does not allow us to observe which governance strategies were used, the documents allow for tracking change and development over time (Bowen 2009, Macnamara, 2005). No current or later developments taint the memory of developments in the crisis, allowing for an unbiased assessment of what governance strategies were used at that time. Due to the political sensitivity of the topic (e.g. during data collection and analysis, a parliamentary inquiry¹ about the governance of PPE procurement was being prepared) other research methods, such as interviews or surveys, were impossible to conduct or unreliable due to bias. Finally, the availability of distinct types of documents serving different audiences and providing different perspectives allowed for triangulating the results.

For this study news articles, parliamentary documents and internal communications of civil servants and public buyers from January 2020 to December 2021 were collected. This timeframe captures the moment right before the first case of COVID-19 was identified in the Netherlands, until the moment when the Netherlands had a) managed to create a stockpile of medical supplies large enough to deal with several uprisings of the virus and b) the outbreak of COVID-19 in the Netherlands appeared to be under control and the government strategy no longer focused on preventing the virus from spreading. For data collection, the two principal researchers developed a data collection strategy. The data was subsequently collected in three rounds between November 2021 and February 2022 by a research assistant. A log was kept describing the collection and cataloguing of the collected data. In total, 973 documents were collected (see Table 1). News articles and parliamentary documents were retrieved from various online archives. Numerous internal documents regarding the procurement of medical supplies during the COVID-19 pandemic were released under the Dutch Open Government Act [Wet Openbaarheid van Bestuur]. This act required governments and public organizations to disclose information if actively requested by the public. Thousands of internal communications were bundled in a PDF file of >200 pages of memos and messages between civil servants from Dutch ministries, Dutch Embassies, potential suppliers, and the consortium (LCH) that managed the procurement and distribution of PPE during the crisis. A supplementary search on Google resulted in the identification of five additional relevant documents, including articles from a procurement magazine in which journalists interviewed healthcare buyers on the PPE procurement during COVID-19. A total of 451 documents were excluded from the original dataset because they were duplicates or, upon examination, did not discuss PPE procurement during the pandemic. Each of the 522 remaining documents (available upon request) was assigned a unique document number, starting with the letter D (e.g. D12). The order of the document numbers is based on the order in which the documents were uploaded to Atlas.ti.

Table 1. Collected data in the various rounds.

	News articles	Parliamentary documents	Other documents
Search terms for all rounds	procurement protective equipment; shortage protective equipment; procurement face mask* covid; shortage LCH covid.	procurement protective equipment; shortage protective equipment; procurement face mask* covid; shortage LCH covid.	procurement protective equipment; shortage protective equipment; procurement face mask* covid; shortage LCH covid.
Retrieved from:	Nexis Uni Nos.nl NU.nl Websites newspa-pers: Volkskrant, Trouw, NRC, AD and Telegraaf	Tweedekamer.nl Rijksoverheid.nl	wobcovid19.rijksoverheid.nl Google
Round 1. (Jan. 20 – Oct. 21)	661)	293	6
Round 2. (Nov. 21 – Dec. 21)	8	2	-
Round 3. (Final check)	3	-	-
Total collected documents	672	295	6
Excluded documents (duplicates etc.)	341	110	0
Remaining documents for analysis	331	185	6

Data coding and analysis

To analyse the data, all documents were uploaded to Atlas.ti. Due to the substantial number of documents, the researchers organized the data based on the month of publication and document type in Atlas.ti. This helped maintain an overview of the data and allowed the researchers to identify patterns over time and across document types (e.g. in- or decrease in news coverage of the topic).

To code the data, a five-step hybrid strategy was used. First, the sensitizing concepts (see Bowen, 2006) from the literature were used to develop a preliminary codebook, with some generic codes for relational and contractual governance strategies, as well as codes regarding the characteristics of the involved actors. Second, the codebook was refined using a more inductive strategy. Themes from the data were clustered and linked to the aggregated concepts defined in the first step. Third, the codebook was tested on a small sample of documents by the principal researchers, and the research assistant was trained to code the data accordingly. Fourth, to ensure a reliable coding process, a calibration session was organized in which the principal researchers and research assistant all coded the same sample. The coding was subsequently compared and discussed, resulting in an additional explanation of the codes. Finally, the research assistant, under supervision of a principal researcher, coded all documents in Atlas.ti using the final codebook (see Table 2), kept a log, and reported on a regular basis. The second principal researcher checked all codes.

Table 2. Final codebook (after test and calibration session).

Sensitizing concepts	Dimensions	Indicators
Deductively derived from literature	Deductively derived from the literature, supplemented, and corrected by inductive coding.	Inductively derived from coding documents
Contractual governance	Contract	Use of contract Agreements in contract on, e.g. on payment, monitoring, quality assurance, what, when and how to deliver
	Screening	Prior to agreement screening of potential suppliers Certificates Quality check products
	Sanctions (formal)	Penalties as agreed upon in contract
	Communication (formal)	Formal letters Administrative procedures Transparency
Relational governance	Joint problem solving	Buyer–supplier aim to help each other Goodwill Diverge from procedures to realize PPE delivery
	Information sharing (voluntary)	Not enforced by contract Sharing information about procurement opportunities
	Good relationships	Gut feeling about partner Trust Commitment
	Flexibility	No hard agreements, on for example: supply, penalties, quality of produce, time of delivery, procedures
	Communication (informal)	Ad hoc texts Calls E-mails Communication via mutual acquaintances
Characteristics of suppliers	Well-known supplier
	New supplier	With expertise: offered PPE on the market before Without expertise: new player in this market Good reputation: well-established business Bad reputation: ‘cowboy’

The coded data was analysed using the phases of the procurement process as described in section two. The coded data was analysed thematically, focusing on which governance mechanisms were employed during each phase of the procurement process and the consequences of the used governance mechanisms. A timeline of the COVID-19 pandemic in the Netherlands was drafted to illustrate the developments over time. Because PPE procurement in the Netherlands consisted of multiple procurement processes that took place at different points of time throughout the pandemic and the research’s unit of analysis is the sector level, the various phases of the procurement process refer to different moments in time. The timeline therefore only discusses the relevant context to the governance mechanisms discussed afterwards. The timeline and the results of the analysis are discussed in the next sections.

Results

Public procurement during the crisis: a short timeline of PPE procurement in the Netherlands

The first COVID-19 case in the Netherlands was confirmed on 27 February 2020 (Rijksoverheid 2021a). It became an official pandemic on 11 March 2020 (Rijksoverheid 2021b). At that time, there was already a global shortage for PPE. A first indication of procurement under pressure was a complaint by the Dutch Dentist union that they were experiencing supply problems with PPE and asked for centralized procurement by the government (D540). Via joint problem-solving and exchanging supplies with other healthcare institutions, organizations attempted to resolve the shortage themselves. They also organized procurement via the existing regional network for acute care (Regionaal Overleg Acute Zorg – ROAZ) rather than at the level of individual hospitals (D56). Simultaneously the Ministry of Health, Welfare and Sport (VWS) started procuring PPE, which is unusual as they had never done this before (D465). By the end of March 2020, the procurement of PPE is centralized nationwide with the establishment of the LCH (Landelijk Consortium Hulpmiddelen, in English: National Consortium Equipment) (LCH, 2022). The healthcare sector initiated the establishment of the LCH, with support from VWS. The LCH is no legal entity nor part of the government, but a consortium comprising experts from healthcare providers, manufacturers, and distribution- and transportation companies and aimed to guarantee the availability of PPE and emergency supplies for healthcare professionals through central procurement, management, and distribution (D111). It struggled initially (April 2020) as various buyers and suppliers displayed opportunistic behaviour. The market was full of ‘cowboys’. Some countries prevented the export of PPE or tried to buy it from under the noses of other countries (D36; D50). The Dutch government therefore also stimulated the local production of PPE. Communication released under the Open Government Act shows a lot of informal contact between the Dutch Ministry of Foreign Affairs and Dutch embassies in PPE production countries between March and May of 2020. These embassies, for example in South Korea, used their local business networks to identify leads to PPE and forwarded offers to the LCH (D573). By May 2020, the scarcity of PPE decreased. Most healthcare institutions had sufficient supply and consequently the urgency for procuring PPE lessened. Between May 2020 and May 2021, local production of PPE took off, resulting in millions of PPE being produced in the Netherlands (D126; D141). In October 2020, the Dutch Central Audit Service examined the procurement of PPE. They reported various accountability issues, such as a bank guarantee issued without formal approval from the Minister of Finance (D477). Besides this report, accountability issues were scarcely addressed until May 2021, when the Dutch court of Audit published a critical report. At the same time, it turned out that a well-known Dutch entrepreneur and political activist who had closed a supposedly ‘not-for-profit’ multi-million PPE deal had made millions in profit. Both events gained much attention in the media and the Dutch parliament (D32; D183). An official investigation into the ‘face mask deal’ by the aforementioned entrepreneur remained in the news until the end of 2021 (end of data collection). In Table 3 the timeline is written down. In the next section, the use of governance strategies is discussed.

Table 3. A brief timeline of PPE procurement during COVID-19 in the Netherlands.

Moment in time	What happened
End 2019	COVID-19 broke out and became a global pandemic
27 February 2020	First confirmed COVID-19 case in the Netherlands
11 March 2020	COVID-19 is officially pandemic
March 2020	Dutch procurement of PPE via healthcare institutions and regional ROAZ network. Organizations share supplies when needed. Ministry starts PPE procurement.
End March 2020	Establishment LCH
April 2020	LCH struggles to procure sufficient PPE
March – May 2020	Dutch ambassies get involved in PPE procurement
May 2020	PPE scarcity decreased
May 2020 – May 2021	Local production PPE in the Netherlands takes off
October 2020	Dutch central audit services reports accountability issues in PPE procurement
May 2021	Critical report regarding accountability issues by the Dutch court of Audit. Word spreads that well-known entrepreneur made millions in profit despite claims to provide PPE not-for-profit.
May 2021 – End 2021	Planned official investigation regarding PPE deal gets lots of news coverage.

Analysis: the use of governance mechanisms during procurement under pressure

In this section, the use of governance mechanisms is discussed for each of the stages of the procurement stages of prepare, purchase and perform. Table 4 shows the most prominent governance mechanisms as identified from the data. All observations of governance mechanisms are discussed at the sector level. This allows for the identification of patterns in the use of governance mechanisms across PPE procurement in the Netherlands. The sensitizing concepts of contractual and relational governance presented in the theoretical chapter are considered a starting point for the analysis.

Table 4. Governance strategies per procurement phase.

Procurement phase		Governance mechanisms
Prepare	Exploration	Informal and direct communication between buyer – supplier (relational governance) Joint problem-solving between buyers of different organizations (relational governance) Joint problem-solving by Ministry (relational governance)
	Initiation	Formalized joint problem-solving via LCH (relational governance) Formalized information sharing via LCH (relational governance)
Purchase	Tender	Informal communication to find PPE suppliers via networks (relational governance)
	Assessment	Assessment of quality and reliability of PPE and supplier using formal criteria, such as certification (contractual governance) Assessment of quality and reliability of PPE and supplier via networks, such as on-site quality check and embassy networks (relational governance) Compilation of list of trustworthy suppliers based on earlier experiences (relational governance)
Perform	Implementation	Use of (incomplete) contracts (expedited and attenuated contractual governance) Standard procedures replaced by emergency procedures (expedited and attenuated contractual governance)
	Execute & extend	Impossibility to use contractual governance and relational governance, resulting in the absence of governance mechanisms (laissez-faire governance)

Prepare: exploration and initiation

In the first phase of the procurement process, the need for PPE is explored. During the crisis, the sudden increase in demand led to informal communication between procures and suppliers, which is considered relational governance. Buyers called well-known suppliers directly to explore the supply of PPE. Various documents also report buyers reaching out to each other to explore if they have PPE in stock, rather than announcing a tender online. Illustrative is the remark of a buyer in a document dated from March 2020: ‘Fortunately, the cooperation between the hospitals in this region is good. If someone within our ROAZ (Regional Consultation Acute Care Chain) was short on surgical masks, there has always been another hospital that still had a batch’ (D52). Via joint problem-solving, another aspect of relational governance, buyers tried to prevent shortages. Initially, information sharing and joint problem-solving happened between individual healthcare institutions or within the care region, but the Ministry of VWS, however, quickly became involved: ‘Under normal circumstances, the ministry has no role in the purchase of these items. Healthcare institutions do this themselves. Now we had to step in, but it took time to find our way around. We first had to map out the demand because that required an overview’. (D17).

In the initiation phase of procuring PPE, it became apparent that the demand of healthcare institutions could not be supplied via contracted or well-known suppliers or solved by sharing PPE within the region. It needs to be procured from other suppliers via the market. Joint problem-solving became formalized in the establishment of a national consortium for PPE, the LCH. The then minister of medical care explained in a news article in March of 2020 (D10): ‘All hospitals do their own purchasing; everyone is looking for face masks. That can be better, right? We take charge, in close consultation with healthcare professionals. It’s not like the ministry says how it should be done. We bring people together. For example, we now purchase scarce items nationwide. If there are any problems, we will cut the knot’. This ministry had never procured such goods themselves or taken on a coordinating role in initiating the procurement of PPE. Later, local production of PPE by reputable local business was stimulated by the ministry (D106). This is an example of joint problem-solving, pointing towards the use of relational governance mechanisms to govern the relationship between buyer and supplier during the initiation phase of the procurement of PPE.

Purchase: tender and assess

While normally the tender phase consists of a highly structured contract-based tender procedure, during the pandemic this phase was much more informal and relational, especially at the beginning of the crisis. The search for suppliers of PPE went both publicly via the media and behind the scenes via formal and informal networks, thereby often relying on relational governance. The internal communication between the LCH, VWS, and various Dutch embassies shows a large degree of informal communication, emails, and calls to forge relationships with potential new suppliers rather than the use of contractual governance mechanisms. For example, Dutch embassies in PPE production countries, such as Morocco or South-Korea, reached out to their networks to spread the word that the Netherlands was looking to buy PPE as illustrated by the following quote of 23 March of 2020. An employee of the Dutch Embassy in South Korea wrote to an unknown contact: ‘Hi [name blacked out], this is the list of items that the Netherlands wants to buy on a short notice. Please let us know if there is any company in your network who could supply this’. (D573). This informal communication is a clear example of the use of relational governance mechanisms in the tender phase.

The combination of urgency, scarcity, and high demand resulted in a complicated assessment phase. Various documents reported screening difficulties, also due to the overwhelming number of people that were trying to help. The reaction of Ernst Kuipers, chair of a large academic hospital, in the media is illustrative (D84): ‘The face mask market has turned into a wild west sector in a few weeks. We get a lot of offers, says Kuipers. At the Erasmus Medical Centre sometimes a hundred e-mails or telephone calls per hour. There were many fraudulent providers among them’. The then minister of medical care shared a similar story in another news article (D17): ‘Acquiring PPE was going to take up a lot of my time. Everyone wanted to help. As a ministry, we had to learn to deal with that. My phone rang all day too. People from my circle of acquaintances, or friends of friends, or friends of friends of friends: they all thought they could get stuff’. The LCH assessed all potential suppliers (D386): ‘The LCH has always assessed on the criteria of price, quality, and security of supply’. Documents show that, for example, products needed to be certified. The requirement for certification is a clear example of contractual governance mechanisms in the assessment phase. However, in practice, assessment using these criteria is complicated. In some cases, PPE was procured that upon receipt did not meet the quality criteria related to the certificate. Therefore, additional relational governance mechanisms were used to assess the quality of the PPE and reliability of the suppliers. LCH and VWS often use informal communication and information from their network partners to do so. For example, the Dutch Embassy in South Korea made use of the non-profit Korean Trade-Investment Promotion Agency (KOTRA). Suppliers on a list by KOTRA were expected to have a good reputation and thus not expected to display opportunistic behaviour. Also, after the initial PPE purchases, the LCH compiled a list of suppliers they now had experience with and had proven themselves dependable and with a good reputation. With these practices, LCH attempted to adhere to existing procurement procedures and contractual governance mechanisms as much as possible. If this was not possible or sufficient, a more relational alternative was used, like relying on existing networks to assess reliability and quality.

Perform: implement, execute, and extend

In the implementation phase, regular procedures indicating contractual governance were generally followed as much as possible. In the data reference is made to contracts being drafted, payments made, and receipts issued. However, especially in the beginning of the crisis, standard procedures were replaced by emergency procedures. This is illustrated by the following written answer to a formal question of a member of the House of Representatives regarding emergency payments (D403): ‘From 1 July 2021, it has been decided to no longer approve so-called emergency payments. In addition to the regular payment method, emergency procedures were also carried out in 2020. Such emergency procedures were sometimes necessary in the acute crisis, where there was more demand than supply, to be able to order protective materials and make immediate payments. However, for normal business operations, this is a difficult flow to control, precisely because it goes outside the regular administrative process’. The emergency procedures allowed for relational governance mechanisms, which were considered necessary to deal with the uncertainties of the pandemic. Various news articles quote buyers, civil servants, and suppliers talking about the uncertainty they experienced. The remark of a local entrepreneur in a news article is illustrative in this respect (D4): ‘It was all a lot of hassle. Contacts with the Ministry of Health, Welfare and Sport and the national consortium [LCH] were quite difficult. But now I can understand that. It was a very strange time. We were all learning. Nobody knew exactly how to do it: not us and neither did the LCH, which was formed with delegates from different companies. We had no experience with face masks. When they googled me, of course they also thought: is this the person to set up such a company? We presented our business plan at the beginning of April, the first machine arrived on April 23, and we received a supply contract from the LCH at the beginning of May. In hindsight, that’s pretty quick’. This quote shows regular procedures, such as the use of supply contracts, as well as deviations thereof, highlighting the continued desire to use contractual governance and how the pressures of the crisis required expedited and attenuated contractual governance.

After contract close the contract needs to be executed and possibly extended. Contract execution was extremely uncertain during the COVID-19 crisis, and opportunistic behaviour was frequently reported in the news: (D182): ‘The parking lot near the juvenile detention centre in Lelystad offers little shelter from the wind and the surprisingly low temperature. […] This is the gathering place for one of the most daring operations in the corona crisis. It is not without reason that the Special Interventions Service (DSI), an elite unit of the police, has been called in. Five motorcycle cops are also on standby to protect against possible robberies by criminals. Four empty trucks park nearby and wait for a signal to be fully loaded. A toxicologist will come along to perform tests. […] A day later, the bitter truth dawns: the five million face masks probably never existed’. While not receiving PPE that was paid for is the most extreme case of opportunistic behaviour from suppliers, quality issues were also frequently reported. Toxicological tests frequently resulted in the rejection of PPE: (D78): ‘The Ministry of VWS has rejected 308.5 million medical protective equipment purchased by the government. It concerns face masks, glasses, aprons, coats, and gloves. A spokesperson for the ministry confirms these following reports from de Volkskrant’. Next to opportunistic and fraudulent behaviour of suppliers, export bans on PPE from producing countries and limited flight availability hampered the import of PPE. These situations go beyond the scope of influence of both relational and contractual governance. Even in the case of a solid contract with well-known suppliers an export ban or lack of flights could still mean a procurement would not be executed. Interestingly, the documents mention neither the use of relational nor contractual governance mechanisms in this stage of the procurement process. There appears to be little or no governance during the execution of the agreed PPE deals. From the documents we learn that governance decisions earlier in the procurement process affected the possibilities for governance later in the procurement process. For example, the decision to use incomplete contracts or skip regular procedures (both part of contractual governance), could make further use of contractual governance difficult later in the procurement process. The following quote from a news article illustrates this: “The ministry cannot provide receipts for goods delivered. More than 328 thousand of Halma’s face masks are lost. ‘According to Halma, these have been delivered, but they do not appear in LCH’s records’. The lawyer of the ministry wrote in an internal memo about the not so strong legal position of the government: ‘We therefore propose to give Halma the benefit of the doubt with regard to these 328,796 masks”. (D125). This illustrates how earlier decisions regarding the governance of PPE procurement – in this case the letting go of standard administrative procedures – prevented the use of contractual sanctions. Relational governance mechanisms were inadequate to govern behaviour ex post, as it requires a good relationship to solve the problem jointly, which in the case of non-delivery or quality issues were often absent. The Dutch Central Audit Service and Court of Audit identified major accountability issues due to the lack of contractual governance in the initial stages of the procurement process: ‘Contrary to regular tender regulation, (major) purchase agreements have been made directly with suppliers. Recording thereof in contracts is not always available at the time the agreements were made. For some purchases, this this was done ex-post. […] Many agreements are not signed by both parties, which puts VWS at risk in the event of a legal dispute’. (D477). This means that in the extend phase of the procurement of PPE a lack of governance earlier in the procurement process caused issues in the execution and evaluation of the procurement. The buyer cannot use further contractual mechanisms, such as legal action, to hold suppliers accountable and govern their behaviour ex post.

Discussion

Procurement in a public sector context in Europe leans towards the use of contractual governance (Grandia 2018). While relational governance is generally considered more suitable for dealing with unexpected situations (Davis 2007), little was known about the governance of buyer-supplier relationships in turbulent times. By analysing news articles, parliamentary documents, and internal communication between the government, embassies and the LCH on the procurement of PPE during the COVID-19 crisis in the Netherlands, we found that during a crisis governance strategies shift. Particularly in the initial stages of the procurement process we identified a brief shift from contractual to relational governance strategies. The pressures of the crisis triggered the use of relational governance strategies, such as informal communication and joint problem-solving, to explore the need for PPE, initiate a purchase, and assess the offers. In line with the literature (Davis 2007), contractual governance strategies were found too slow and inflexible given the volatile market situation. As not fulfilling the need for PPE would put lives at risk, it is not surprising that we also found that during a crisis ‘public procurement rules get out of the way to free up public buyers to do all they can do get the required supplies’ (Sanchez-Graells 2020, 82). Nevertheless, as already stated by Folliot Lalliot and Yukins (2020), this does come with ex-post risks, as governance choices, including the use of very incomplete contracts in the first procurement stages, can cause laissez-faire governance in the later procurement stages, curtailing the ability to to govern buyer-seller relationships either contractually or relationally and giving rise to accountability issues. The findings indicate that relational governance mechanisms were mainly employed in existing relationships with well-known suppliers. Business supply chains studies explain how such a combination of past experiences and the ‘shadow of the future’ enables the successful use of relational governance (e.g. Bonatto, Martins de Resende, and Pontes 2020; Lumineau and Henderson 2012). While it might seem enticing for actors to exploit a contract in the short term, partners would hardly consider them a trustworthy partner for future contracts due to such actions, consequently reducing opportunistic behaviour (Brown et al. 2007). Likewise, our study reveals that in situations where the relationship between buyers and well-known suppliers was expected to continue after the crisis, little opportunistic behaviour could be identified in response to the use of relational governance strategies. Conversely, when dealing with unfamiliar suppliers and new market players, the absence of an existing relationship left no basis for trust. The data indicates that actors found relational governance strategies insufficient to prevent opportunistic behaviour, causing them to attempt to shift back to contractual governance. This aligns with findings from business studies that found that in case of environmental uncertainty, contractual governance provides greater viability, control, and coordination than relational governance (Bonatto, Martins de Resende, and Pontes 2020; Pilbeam, Alvarez, and Wilson 2012). Moreover, the market situation where repeated transactions were not expected, transformed procurement into a ‘one shot-exchange’ heightening the risk of opportunistic behaviour. A higher frequency of transactions leads to increased dependency, which, as noted by Bonatto et al. (2020), exerts a moderating influence on performance through its impact on trust. The establishment of the LCH brought about the possibility of multiple exchanges, which created an institutionalized form of joint problem-solving by uniting all relevant public and private actors within a consortium, ensuring the availability of PPE for Dutch healthcare professionals. Joint or collaborative procurement has been shown to yield benefits in various other contexts, including cost savings, knowledge sharing, economies of scale, and enhanced resilience (Carrera, Katik, and Schotanus 2021; Walker et al. 2013). However, the return to contractual governance during the crisis did not mean going back to the exact same contracts and procedures that were used pre-covid. The government’s top priority was acquiring as much PPE as quickly as possible to save lives, with other factors like transparent procedures or price taking a back seat (VWS 2020; D364). Consequently, procedures were modified, and contracts were incomplete, sometimes even without a signature. The finding that in times of crisis attenuated contractual governance is used when there is also no solid foundation for trust and relational governance, appears closely related to monitoring and accountability issues that were identified in later stages of the procurement process. The lack of clear (contractual) agreements made governance in the final stages of the procurement process challenging and created governance capacity deficit. Our most surprising finding, however, was that sometimes a conscious decision was made to not govern at all. In keeping with leadership literature (e.g. Eagly, Johannesen-Schmidt, and van Engen 2003) where laissez-faire refers to a general failure to take responsibility, we refer to this as ‘laissez-faire’ governance. The use of laissez-faire governance affected the possibility of using contractual governance strategies later on. For example, if the delivery of supplies is not monitored, then a sanction for the delivery of poor-quality supplies or failure to deliver cannot be administered due to lack of evidence, resulting in further non-governance. The use of ‘laissez-faire’ governance strategies in times of crisis is perhaps unsurprising, as time constraints and societal or political pressure may necessitate shortcuts. However, it does matter what kind of shortcuts are taken. Poppo and Zenger (2002) and Warsen et al. (2019) showed that contractual and relational governance are complementary and that the right mix of relational and contractual governance strategies can have a positive effect on the success of a project. As complements, growing levels of trust might allow for less complete contracts, while strengthening contracts might help safeguard the transaction when trust deteriorates (Brown et al. 2007). However, in times of crisis, combining contractual and relational governance is both inevitable and challenging. Time to install elaborate contractual governance mechanisms is lacking, enhancing the need for relational governance. However, time to build strong relationships that form a solid base for the use of relational governance mechanisms is absent too. This may lead to a risky combination of incomplete contracts and no relationship to moderate the performance, and thus laissez-faire governance. It appears the right mix of relational and contractual governance is key to a successful procurement project, and the right mix of relational, contractual, and laissez-faire governance is needed during a crisis to reduce accountability risks and enable ex-post governance. It is therefore interesting that during the crisis there was little to no attention for governance or lack of governance in the media, parliamentary documents as well as internal communication.

Conclusion and limitations

By answering the question ‘Which governance mechanisms are used in the public sector to govern the buyer–supplier relationship during a crisis?’ our paper contributes to a call for more insight into governing buyer–supplier relationships in times of crisis (McEvoy and Ferri 2020; Phillips et al. 2022; Rosell and Saz-Carranza 2020). The findings show a brief initial shift to relational governance in the exploration, initiation, and tender phases, followed by an attempt to return to contractual governance, albeit with incomplete contracts and attenuated procedures in the assessment and implementation phase. In the literature relational governance is considered a more flexible governance alternative to contractual governance, making it more suitable for governing buyer–supplier relationships during a crisis (Granovetter 1985; Poppo and Zenger 2002). The use of relational governance mechanisms in the procurement of PPE during the COVID-19 crisis, such as informal communication with well-known suppliers and joint problem-solving, on the one hand, and the highly incomplete contracts, on the other hand, support this. However, we also found that the preconditions to make relational governance strategies successful were also absent during this crisis. The conclusion that during a crisis the preconditions for both relational and contractual governance can be absent, is an important theoretical contribution of this study. We note that the need to procure PPE, namely, to prevent healthcare professionals from succumbing to COVID-19, was all-consuming. The awareness that lives were in jeopardy, combined with extreme market conditions resulted in a situation where neither contractual nor relational governance mechanisms were possible or effective. The global scarcity of PPE necessitated one-shot exchanges with unknown and often new suppliers, limiting the possibilities and effectiveness of relational governance mechanisms, while time constraints led to incomplete contracts and expedited procedures, similarly constraining the possibilities and effectiveness of contractual governance. We therefore conclude that during truly turbulent times, when lives in the balance, neither contractual nor relational governance is suitable for governing buyer–supplier relationships, resulting in laissez-faire governance.

While understandable under the pressures of the crisis, laissez-faire governance can cause ex-post risks, such as accountability and monitoring issues. For instance, we observed that improper administration upon receipt of the PPE resulted in the inability to use contractual governance mechanisms ex-post, creating accountability issues. This illustrates that the choice of a particular governance strategy, or the absence thereof, at the outset of the procurement process can lead to a deficiency in governance capacity later in the procurement process. Contrary to the advice provided by Ansell et al. (2021), we conclude that in the case of crisis procurement there is no need for more robust governance strategies; instead, there is a need for clear prioritization. Actors need a prioritization of what aspects in the procurement process and with whom must be governed in turbulent times, and what can be ‘non-governed’. Prioritization enables actors to govern the buyer–supplier relationship at least partly during all stages of the procurement process, limits accountability issues and should be an integral part of any crisis procurement or pandemic preparedness policy. It also facilitates ex-post contractual and relational governance. Therefore, we recommend further research into prioritization and laissez-faire governance to examine how actors determine what is most critical in the procurement process, the conditions under which laissez-faire governance emerges and the overarching considerations that underpin governance strategies.

Regardless of its contributions, this study also has its limitations. First, our decision to do a document analysis limits the kind of information we were able to retrieve. Governance strategies are not only written about but also reflected in the behaviour of actors. Document analysis does not allow for the observation of behaviour and therefore does not provide a complete overview of all used governance strategies. It also does not uncover all underlying reasons and motivations of actors to use (contractual or relational) governance mechanisms. Supplementary interviews would have addressed this, but the continued political sensitivity of the topic and the ongoing parliamentary inquiry prevented this in this study. However, the document analysis does provide relatively objective information, as the contents of these documents are not tainted by later events and issues (like the vast media attention for the famous Dutch entrepreneur profiting from PPE sales to the government under a not-for-profit guise). Second, the scope of this paper is limited, as it only examined the governance of buyer–supplier relationships in the procurement of PPE during the COVID-19 crisis in the Netherlands. Although similar to other research, we did observe that the PPE shortage put significant stress on procurement during the COVID crisis (Arrowsmith et al. 2021; Finkenstadt and Handfield 2021) and that governance decisions during crises can lead to transparency and accountability concerns (Dikmen and Çiçek 2023; Telles 2022), we cannot generalize the findings from this study directly to other countries, contexts, or crises. Because this study shows that context matters for the use of governance strategies, we recommend further research into the governance of buyer–seller relationships during different crises or in different contexts or countries, to see if similar governance strategies, or lack thereof, are employed in other situations where procurement is put under pressure.

Acknowledgments

We would like to thank Pascal Godee for his help in collecting and coding the documents.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was financially supported by the 25-25 programme of the Erasmus University of Rotterdam.

Notes

1. The parliamentary inquiry is the most powerful instrument the Dutch parliament has at its disposal to perform its duty to scrutinize the work of the Government.