On the ex-ante evaluation of Green Public Procurement criteria

Motivation

GPP is defined as a process whereby public authorities seek to procure goods, services and works with a reduced environmental impact throughout their life-cycle when compared to goods, services and works with the same primary function that would otherwise be procured. Unpacking that definition helps pin down many important characteristics of the information gathering and processing that must take place to support this major sustainability transition process. A most striking high-level aspect is the vast scope of this process: With public procurement forming anywhere from 10% to 30% of an economy, there is vast variability in underlying economic activities, environmental impacts, market characteristics and procurement procedures. Nevertheless, the definition of GPP demands a certain degree of coherency across this vast domain, not least to ensure the cost and energy of public resources dedicated to GPP are effective in achieving their target.

Our specific focus here is on GPP criteria and policies, and whether and how they can be evaluated in a consistent way across the board. Before commencing with that discussion though, let us bring to the foreground the hidden assumptions of the GPP definition. These include:

• an assumption that standardized, universally agreed indicators of negative environmental impacts (externalities) from all relevant economic activities (production and consumption) are available (at least whenever those impacts are material). In particular these indicators do not leave any blind spots. For example, they cover both direct and indirect (e.g., upstream and downstream) impacts
• an assumption that such indicators are applicable (via suitable mappings and adaptations) specifically to public procurement activities
• an assumption that reasonably mature markets exist for goods, services and works with reduced environmental impact (hence procurement authorities have options to exercise)
• that procurement authorities engage in informed target setting for reducing their impact as measured by said indicators, but potentially in tandem with other objectives (multidimensional optimisation)
• that environmental targets are reflected in the ex-ante compilation and evaluation of tenders and contracts (via a variety of policies and criteria)
• that the efficacy of impact reduction processes can be verified ex-post on the basis of awarded contracts and completed activities

While significant efforts are underway worldwide to materialize the above blueprint and thus utilize the leverage of public sector procurement towards achieving sustainability goals, the transformation process is arguably still very far from complete. In this blog post we discuss a very specific challenge: how can the wide range of possible policies (collectively sometimes denoted as Green Public Procurement (GPP) criteria) can be approached from the perspective of consistent ex-ante evaluation of their impact.

What is the optimisation objective?

As mentioned, we focus here on how to assess the impact of GPP policies on environmental impact targets. Yet the selection and implementation of GPP policies is likely to be jointly optimising a large number of other considerations, such as:

• the budgetary importance of a sector
• political priorities
• cost considerations and practicality
• visibility to constituents and potential to influence markets

Via the monetization of externalities one may in theory integrate financial and environmental dimensions but here we assume public authorities are in the first instance aiming to manage their environmental footprint as such.

Optionality in Public Procurement GPP choices

We sketch first the variety of optionality that is in-principle available to public procurement authorities when they implement a GPP program. This list is not complete, e.g. there are different possible procurement procedures (open, restricted etc.), we focus on the most directly influential choices.

To start with, approaches towards GPP policies can range from:

• not being required / implemented at all
• being voluntary
• being mandatory but only for some product groups or for some classes of purchasing authorities
• being fully mandatory across the board

This variety of attitudes has an important consequence: procurement activities that do not fall under any GPP policies must also be tracked for their impact in a consistent manner. The ex-post attribution of impact to ongoing procurement activities must be comprehensive for the overall target setting to be meaningful.

Besides the discretionary scope GPP policies, procurement authorities have also wide discretion on the nature of such policies. For example, the EU GPP criteria include two levels for each sector covered:

• core criteria that are designed to allow easy application of GPP, focus on the key areas of environmental performance of a product or service, and aim to keep administrative costs for companies to a minimum.
• The comprehensive criteria that take into account more aspects or higher levels of environmental performance, and are for use by authorities that want to go further in supporting environmental and innovation goals.

Further optionality involves the ability to choose between:

• modifying the specifications of existing production or consumption practices (incremental changes)
• the adoption of new technologies and systems (step changes)

Simply put, aspiring 100% coverage of public procurement by GPP does not quantify the impact reduction, this will only follow from examining the policies. It is thus useful to sketch what are currently the main categories of GPP policies.

Classification of GPP policies

Selection Criteria

Selection criteria in this context concern the profile of economic agents (sellers, contractors) that are eligible to act as counterparties to procurement tenders, effectively reducing the scope of the pool of such counterparties with which the purchasing authority is willing to trade. Credential requirements can be established e.g., via a proven track record in using new technologies or approaches, having relevant certifications etc. This class of criteria exerts only indirect influence on procurement impact outcomes. They represent soft factors, in contrast with the hard factors that can be directly reflected into environmental indicators. Hence, we do not include them in the current discussion, even though they can be of great significance in practice: The complexity and immaturity of the GPP processes means that relying exclusively on imperfect hard factors makes the process more exposed to gaming.

Title, Subject-Matter and Technical Specifications

These attributes define explicitly desired environmental characteristics and performance requirements of the goods, services, or works being procured. The title and subject-matter of a contract is about what product, service or work is desired. The choice of subject-matter is particularly important because it determines the permissible scope of specifications and other criteria that apply.

Specifications are where the quantifiable impact reductions can be introduced. Within the limitations of legal procurement processes they can address any desired set of sustainability indicators (including indirect supply chain aspects). In current practice specification form of a type of minimum requirement. Inclusion of such technical specifications effectively reduces the universe of eligible goods and services (within a certain equivalence range). For example, minimum energy efficiency requirements (e.g. defined via ecolabels) for office equipment, vehicles and buildings helps reduce the impact of ongoing (downstream use) of these facilities.

For this class of criteria, even on an ex-ante basis a certain amount of environmental impact reduction can be asserted (subject to a baseline estimate - see below).

Award Criteria

Award Criteria are used to evaluate bids based on their environmental profile beyond the pass/fail or minimum requirement considerations of technical spec. Award criteria are instead weighing and scoring environmental impact on a scale. Tenders offering better environmental performance can be given more marks. Such point-systems aim to reward incremental performance. This approach can be used to incentivize suppliers to go above minimum requirements.

On an ex-ante basis it is only possible to assert a range of possible impact reductions. For this type of criteria it is only post-award when the actual expected reduction can be confirmed.

Contract Clauses

Contract clauses outline contractor performance requirements that must be met during the execution of an awarded contract. They may include provisions for monitoring and reporting on environmental performance, as well as penalties for non-compliance. For some types of contracts, it makes sense for contracting authorities to look beyond the primary or first-tier contractor, to ensure that environmental requirements will be met also by subcontractors. Such clauses may reduce the uncertainty around the expected outcome. Their aim is to incentivise contractors to deliver at or above expectations. In any case, for our purposes contractual clauses also fall under the category of soft factors. They do not directly modify the impact indicators of GPP policies.

Differentiation in Supply, Services and Works contracts

Reflecting the vast range of economic activities involved in public procurement, there are three major types of contractual relations between the procurement authority and private markets. The type of GPP criteria is adapted to this reality, raising the challenge of a consistent assessment across the board.

Supply Contracts:

Supply contracts deliver concrete goods in defined quantities. In this case GPP policies must address:

• The environmental impact of energy, materials, land use etc. used to make the product (production)
• The impact of the packaging and transportation of the product (delivery)
• The impact during use of the product, including the durability/lifespan of the product (use)
• Opportunities for recycling/reusing the product at the end of life (disposal)

A strong component of supply type contracts concerns impacts that have taken place in the past, in the context of the production of goods that are then delivered. In other words, the embodied impact that takes place in the upstream chain. Under the procurement directives, materials and methods of production or provision can be taken into account when defining technical specifications - even if these do not form part of the material substance of what is purchased, for example electricity which is produced from renewable sources or food produced from organic agriculture.

Service and Work Contracts:

Service contracts accentuate also impacts that occur for the duration of service provision contract, e.g., the energy and water consumed, and waste generated in carrying out the service. Yet products/materials used in carrying out the service may also have an upstream impact component. In addition to the above considerations, works contracts may have significant environmental impacts e.g. in respect of land use or traffic planning.

Resolving Economic Complexity

A procurement project may be a simple transaction that involves e.g., the supply of a single bulk good. In this case the association of a single granular GPP criterion with a corresponding measure environmental impact is likely entirely adequate. But a tendering process (especially large works) may comprise multiple activities that involve a range of impacts, with many of them occurring anywhere else along the supply chain. For large works activities a formal Environmental Impact Assessment might be carried out.

If the procurement involves more than one materially impactful activity, the total impact reductions from any GPP criteria must be resolved to and quantified as the sum of the impact reductions from each underlying activity. Properly identifying and defining the material GPP activities is thus crucial, as impact reductions must be determined separately for each material activity associated with a Procurement Contract.

GPP Activities and their Effects

A GPP Activity is thus a specific action or intervention that is singled out in the procurement contract via GPP criteria. It aims to achieve a measurable Environmental Impact, in other words, have a concrete Effect. GPP Effects are the measurable changes in Environmental Impact caused by a GPP Activity associated with the Procurement Contract.

Broadly speaking, an effect must be measured as a before/after calculation, where the before (or baseline) scenario is the estimated impact in the absence of the GPP criterion and the after is the estimated impact as inferred from the application of the criterion (which may only guarantee a minimum level). The precise mechanics of quantification depends of course on how the environmental impact indicators are associated with the volume and type of economic activity.

Estimating Baseline Impact

A Baseline Impact is a reference scenario that must be juxtaposed to any of the included GPP Criteria / Activities comprising a GPP contract. On an ex-ante basis, it is a hypothetical description (a counterfactual) of what would have most likely occurred (i.e., what type of procurement would have happened) in the absence of any considerations about environmental impact mitigation, i.e., without the application of any GPP Policies.

A reliable baseline impact is obviously required if one is to assess the reduction of impact due to GPP policies. Yet it is not necessarily a uniquely defined metric: There could be more than one realistic alternative baseline scenarios. In other words, a variety of alternative procurement requirements (none of them involving any GPP criteria) maybe be plausible. These may be substantially different as far as environmental impact is concerned. The most appropriate baseline scenario depends on context. E.g. whether there has been prior contracting activity for the same goods or services, whether the market has already migrated to newer technologies or practices etc.

The challenge of Impact Additionality

An emerging difficulty is that many activities that reduce Environmental Impact (relative to historical levels) would happen regardless of the existence of a GPP program. If a project had happened anyway, then attributing the corresponding impact reduction to GPP policies dilutes the monitoring of the effectiveness of the GPP program.

So-called Additionality is a concept that is precisely relevant in distinguishing the genuine effect of a GPP Activity from its Baseline Scenario. It is characterisation applied to GPP activities, stipulating that impact reductions should only be quantified if the GPP activity would have not happened anyway, i.e., that the GPP activity (or the same technologies or practices that it employs) would not have been implemented in the baseline scenario.

Example: Replacing older generations of combustion engines with newer combustion technologies may be driven purely by cost reduction (or even the unavailability of older technologies in the market!). The baseline scenario that produces a fair measure of GPP impact is one that assumes the procurement of newer versions of the same overall technology. In this instance utilizing as a baseline the historical pattern (using obsolete technologies) creates an artificially large impact estimate for the GPP policy.

The devil is in the details

While the overall logic of the GPP policy evaluation scheme is fairly simple and clear, the devil is always in the implementation details.

Across the nexus of IT systems, sources of GPP criteria, contract databases and reference data (environmental indicators) there can be many gaps that need to be bridged. Indicatively some examples:

Multi-criteria labels are the most common type of environmental label and also the most commonly used in GPP. Multi-criteria labels are based on scientific information about the environmental impact of a product or service throughout its life cycle, from extraction of the raw materials, through production and distribution, the use phase, and final disposal. Yet unless the different criteria can also be disaggregated it can be challenging to translate such a label into usable indicators.

Life-cycle assessment (LCA) allows for cradle-to-grave analysis of the environmental impact of products. It thus includes the extraction and refinement of raw materials, manufacturing and other stages of production through to the use and disposal phase. LCC may also include the cost of externalities (such as greenhouse gas emissions). Carrying out an LCA for an individual contract though implies considerable extra effort.