Financing Better Systems: What Financial Institutions Can Do To Build Better Systems
This piece was originally published on Illuminem.
Over the past decade, environmentally cleaner systems have increasingly become the more economically compelling ones. As technologies have grown more efficient, more capable, and cheaper, decarbonized systems have overtaken their carbon-intensive equivalents, because they cost less, perform better, and are more reliable. The largest emissions reductions of the past decade came not from corporate net zero targets but because clean alternatives now often create more value than what they replace. Building efficiency is driven by operating cost; the pairing of EVs with flexible charging by a cheaper, better customer offer; the scale-up of clean power by price and speed. Decarbonization is increasingly a byproduct of building better systems, not the reason for building them. These ‘better’ systems include, for instance:
- Energy and power generation systems producing energy that is more affordable, reliable, resilient, and secure;
- A built environment, integrating efficient, smart buildings, transport and energy infrastructure for reliability and cost savings;
- Industrial and manufacturing systems that integrate advanced materials and processes for competitive advantage, with minimal waste;
- Transportation systems that ease congestion and improve air quality;
- Food and agricultural systems that are more productive and resilient.
A narrow focus on carbon misreads what drives the systems that reduce overall emissions. A battery or grid-component factory in Southeast Asia may have high emissions because it secures energy from a carbon-intensive grid, but it enables decarbonization of the wider system; the same holds for the transmission lines, refining capacity, digital optimization, and storage any clean system depends on. A narrow focus on entity-level emissions, and their contribution to ‘financed emissions’ ledgers, misses the system-wide decarbonization outcomes individual actors or investments enable. Moreover, the adoption of many of these higher-value alternatives – the efficient buildings, flexible energy systems, agtech – are not driven by decarbonization imperatives, nor enabled by them. They are financed and adopted because they create value for their customers and investors, through new markets and revenue streams or cost savings and efficiencies.
Finance flows to what is financeable. Financial institutions are mandated to create and preserve value, each within its own binding constraints and over time horizons typically far shorter than the systems they finance: for banks, capital and credit limits and lending tenor; for an asset manager, mandate and benchmark; for a private equity firm, the hold period and return threshold. They are not mandated to determine what optimal societal outcomes are, nor would anyone want them to be. But that limit does not make finance passive. Within these mandates, and in coordination with public and private actors, financial institutions can find and scale the value that increasingly lies in the better systems themselves and identify where value could be created that no one is yet financing.
When the better system is also the cleaner one, financing it is the thing that most advances decarbonization in the real economy and is consistent with financial institutions’ mandates. That reality can be described the same way to a deal team weighing a transaction, to a shareholder asking about return, and to anyone pressing finance to do more: value is being created in systems that current frameworks overlook, and the work is to find it, make it financeable, and clear what stands in its way.
This paper focuses on climate, because that is where the pressures on financial institutions are most immediate and most contested, and where the frameworks built in response are most misaligned, but as emphasized above, decarbonization is best understood as one characteristic of better systems — alongside access, affordability, reliability, security, efficiency, etc. When the orientation is the system rather than the entity, the connections among these objectives become clear. It builds on a first paper, Redefining Corporate Climate Action, which argues that the relevant unit for corporate climate action is the system, not the entity. A second, From Planetary Hazard to Financial Stability: Disentangling Climate Risk and Institutional Responsibility, separates two things that finance frameworks routinely conflate: building cleaner systems, and managing climate-related risk.
The two are easily confused. Understanding how physical risks are evolving, and how they move through insurance markets and public buffers into asset values, is prudent risk management. Risk teams should track how those risks are intensifying and shifting, and price them accordingly. That work is important, but it is not the subject of this paper. A related argument asks financial institutions to manage systemic risk and not only their own exposure. Better systems do reduce systemic risk, and a transition that is financeable is a smoother one, but that is an outcome of the work rather than something any single institution can be held to delivering: systemic risk is a property of the system, not a target an institution can be assigned. This paper is about the separate question of building cleaner systems.
The frameworks built for financial-sector climate action also entrench the misapplication of ‘net zero’ to mean entity-level carbon neutrality. Financial institutions are asked to neutralize their own footprints, set financed-emissions targets, align portfolios with modeled pathways, and require the same of the companies they finance. None of this acts on the systems that produce emissions. A financed-emissions ledger reports the composition of a portfolio, which largely reflects the economy that portfolio finances, not the institution’s effect on the real economy or its intent to change it. An institution can lower its financed emissions by selling exposure to a high emitter that keeps operating under a new owner, improving the ledger without impacting real global emissions.
The real economy is shaped by many interacting actors: private firms, financial institutions, and a public sector that is itself many distinct actors. A pervasive market versus policy dichotomy – at times suggesting that climate action must be policy-led and other times calling for private sector leadership – over-simplifies the roles of public and private actors. Private capital largely finances what is already financeable, and that universe is shaped by myriad public institutions, policies and instruments, spanning land-use, permitting, approvals, liabilities, procurement, standards, mandates, and other market-shaping regulation.
Changing how systems operate often requires planning, coordination, sequencing and collective action among the public and private actors that shape the system. Yet few actors are able or incentivized to see what a system could deliver if its parts were arranged and financed together, and none can get there alone. This is as true of public actors as private ones. An agency sees the piece it is responsible for, and the people who could work out what a different arrangement would make possible are usually somewhere else, along with the instruments that would move it. Each actor does its own job, which sometimes produces what the system needs but often does not, because getting the actors aligned is costly and because the gains from a better arrangement fall to everyone while the cost of moving first falls to whoever moves. What goes unbuilt is rarely what no one wants; it would often serve everyone, but no actor can achieve it on its own.
Financial institutions largely know what makes an investment financeable, and that knowledge is needed in building these systems. What is missing is not a sharper set of decarbonization targets or more accurate portfolio metrics, but the work of identifying and designing the optimized systems, in energy, transport, the built environment, food, and industry, diagnosing what keeps capital from reaching them, and designing the structures that make them financeable. Rather than seeing each investment or project as meeting a financeability hurdle, a financier sees which constraint recurs across every deal of that kind and what change – of policy, institutions or risk allocation – would allow capital to flow into that whole class of deals, rather than project by project. Financial institutions’ meaningful engagement, therefore, focuses on the specific constraints and enabling levers that would unblock a specific system outcome in the real economy. For example, developing a utility regulatory framework that recognizes flexible industrial load as an asset to the grid, structuring a retrofit market so lenders can underwrite it, or supporting robust fiscal space for catalytic public investments in enabling infrastructure or R&D. The useful question is which actor — a procurement authority, a standard-setter, a regulator, a planning ministry, a rating agency, an international institution— holds the lever that would unlock the financing a given system needs, and what that actor would have to do.
Below, we describe three ways in which financial institutions can help build the systems that underpin a sustainable economy, optimized for cost, affordability, access, reliability, resilience and competitiveness, as much as for decarbonization, and to make those systems financeable. They are ordered from the most self-contained to the most collective: identifying what is already financeable, building the structures that make investments financeable, and informing the reforms that make whole systems financeable.
1. Identifying what is already financeable
The first requires no new instruments and no concessionary capital. It asks only that financial institutions look for value created in ways their frameworks do not capture. Value is increasingly created and efficiencies gained through means that are not traditional projects or clients: efficiency, digitalization, flexibility, and integration that became possible as technology advanced and costs fell. These are often not framed as climate investments at all, and they do not arrive through established pipelines, so seeing them takes deliberate effort. Once demonstrated, many are replicable.
The cost of clean technology has fallen far enough that flexible, distributed energy systems now create value in ways that were impossible a decade ago. Integrated with electrified transport and industry, they open new revenue streams, defer investment, and lower system costs, with the emissions benefit incidental to a design choice made for efficiency or value. Such gains rarely show up on any emissions ledger, because system-level improvements are hard to attribute to one actor and often improve cost or service rather than producing a clean, reportable reduction. The same logic scales, from industrial parks built around circularity to grid-integrated storage, demand response, and materials substitution. A chip whose redesign cuts processing time also cuts energy; battery and grid-component manufacturing underpins the entire transition yet falls outside every climate taxonomy. The value is real, but the frameworks are not looking where it is.
Some of these investments are already financeable on their own merits, yet go unmade for two related reasons. The first is visibility: many new investments are non-traditional and so will not surface in benchmark-driven allocation. Sometimes the value is split across categories that are underwritten apart, as with an integrated system pairing generation, storage, and demand management, whose value comes from the parts working together but whose parts are each financed by a different desk against a different model, so no single mandate covers the combination. In other cases, the asset fits a recognized category but fails the criteria that route capital within it, the sector benchmark, the minimum deal size, the demand for a proven and standardized asset. And in still others, the system or technology is simply newer than the market’s habits: the value is real, but without a track record or an established way to underwrite it, allocators do not yet price it for what it is.
The second reason is that some of these investments are considered riskier than they are. For instance, a technology that is already competitive and on a clear cost-down trajectory, may still be treated as riskier than it is because the technology or category is unfamiliar to the financial sector. What it takes to identify or re-price these investments varies by institution. Direct lenders, specialist funds, and family offices can move on a thesis like this readily; larger pools of capital in benchmark-relative mandates cannot. In some cases, the perceived risk is embedded in official credit ratings, and mandate constraints prevent institutional investors from assessing risk independently of it.
Seeing and repricing these investments reaches only those cases where the constraint is awareness and risk perception because the investment itself is already competitive.
2. Building structures that make investments financeable
The second is to build structures that let a financial institution finance things it otherwise could not. This takes two forms. The first is the structuring that institutions already do as a matter of course, on their own books and within their own instruments, to make a transaction work. The second reaches further, to structures no single institution can assemble alone, where making a project financeable depends on guarantees or risk-sharing from other actors. Both start from investments that are not yet financeable as they stand but could become so, and both stay within the commercial mandate. In practice these run across a familiar set of structures: co-investment and origination-sharing vehicles that let one institution take on assets it could not originate alone; risk-transfer arrangements that move first-loss or specific risks to whoever is best placed to bear them and free up capital for new lending; aggregation and warehousing platforms that pool small or non-standard assets into something financeable at scale; and dedicated credit and guarantee lines that open a class of client or market that standard products do not reach.
Many systems solutions fail to clear institutional thresholds not because long-term planetary risk is mispriced but because discrete, near-term, project-level risks block them: offtake uncertainty, currency and sovereign risk in emerging markets, revenue volatility, first-of-a-kind technology risk, tenor mismatch, demand that has not yet formed, coordination gaps across a supply chain, the absence of coordinated market design. Discounting climate risk into cash flows does nothing for any of these. Financial structuring sometimes does.
Much of this is ordinary practice. Banks, insurers, asset managers, and funds manage and reshape risk every day, through syndication, securitization, collateral and security arrangements, tranching, blended structures, and tailored terms. These tools are routinely used to enable transactions that a single straightforward instrument cannot. The expansion would be applying them to systems they have not yet been used for. By spreading exposure that is too concentrated, matching tenor more closely to the asset, or isolating a first-of-a-kind risk so the rest of the deal can clear, these structures can be decisive in financing systems that would otherwise fall just short. None of this requires a new mandate or outside support, and none of it depends on concessionary capital or below-market return. The economics stay commercial: the tools reshape risk the institution already prices rather than subsidizing a deal that does not otherwise work. What it takes is the effort of structuring, which is the institution’s ordinary business.
Other constraints lie beyond what any one institution can resolve on its own books, and the structuring then depends on other actors. Many of the risks that make compelling investments unfinanceable are perceived, first-of-a-kind, or offtake risks that become manageable once shared. An institution can bring in a guarantor, an insurer, a development finance institution, or syndication partners to take a tranche of the risk, and a guarantee rarely called can move a project across the financeability threshold. These are arrangements an institution can initiate itself, with counterparties and instruments already available to it.
Such arrangements are already emerging. Asset managers and development banks co-invest in origination-sharing models, taking on assets neither would originate alone and giving institutional capital access to a development-finance asset class it could not otherwise reach. Asset managers and insurers pair up so that specific risks are underwritten by the party best placed to hold them, letting capital flow against exposure it would otherwise avoid; in the significant-risk-transfer market, insurers take first-loss positions on slices of bank loan portfolios in exchange for freeing the bank’s regulatory capital for new lending, a market that moved over $1 trillion in loans in 2025. Each pairs an institution that understands a risk with one positioned to bear it, opening a class of investment neither could reach alone.
3. Informing the reforms that make systems financeable
Much of what limits financing lies beyond what any single institution can resolve, because the constraint is market coordination, risk allocation, or a structural failure of financial architecture. This holds whether a system is being built or already running. Greenfield means designing and financing the optimized system from the start. For an economy in motion, being the better system does not by itself retire the worse one, which is already built, contracted, and financed; displacement therefore depends on changing the cost and risk structure of assets already operating, financing the conversion, sequencing the switch, and absorbing the stranded-asset risk that holds incumbents in place.
Consider the risk assessments that make many emerging-market investments prohibitively expensive even when they are sound. Take a utility-scale solar project in a low-income country: the resource is excellent, the technology is proven, and the levelized cost undercuts the fossil generation it would displace. On its own economics it is plainly financeable. Yet it carries a cost of capital several times that of an identical project in a wealthy country, not because anything about the project is riskier, but because it inherits its country’s sovereign rating. Sovereign rating ceilings, controlled by three agencies, cap the rating of almost any entity in a low-income country below investment grade, regardless of the credit quality of the specific project. Some of this is genuine misperception. Some is real risk that other actors, prudential authorities and the governments of wealthy countries among them, could and should address. No single institution can lift a sovereign ceiling, and changing agency methodology is slow. But banks and insurers understand exactly how the ceiling translates into an unfinanceable cost of capital, which is precisely what the actors who could change it often do not. The useful contribution is neither to de-risk these countries one project at a time nor to call in general terms for climate policy. It is to bring the specific obstacle to the actors who can move it. Prudential regulators could permit insurers and banks to look past the sovereign ceiling where a project-level credit assessment shows that the channels through which sovereign stress would reach the project’s cash flows are either inactive or contractually mitigated, and that reform could be carried into the G20 agenda. An individual institution cannot simply disregard the ceiling; doing so raises its measured exposure and capital charge. So the prudential authorities are responsible for the reform, but the institutions that price the risk may understand it better than the regulators who could change the rule. Closing that gap is itself a contribution.
It also means helping design the mechanisms that do not yet exist. A pooled guarantee fund for industrial decarbonization could take the context-specific risks that block individual projects and convert them, in aggregate, into something traditional institutions can finance, and debt issued against such a fund could raise capital at a scale, and with a systemic impact, that labeled issuance cannot. Individual green issuance is marginal by comparison; it may commit a company to what it could already do, but it does not change the system that determines what the company can do. No single institution can stand up a facility like this alone, but an institution that understands what makes the underlying projects financeable is well placed to help design it and to convene the parties who can.
Working alongside the other actors who shape a system is the most direct of these contributions, and the most neglected. Financial institutions are uniquely knowledgeable about what makes a system financeable and which constraints persistently block it. Take hyperscalers. They raise capital at enormous scale and finance themselves without difficulty, and the firms building them are mostly indifferent to how they connect to the grid. Left to itself, a data center draws on existing supply and raises prices for everyone, or builds dedicated generation that does nothing for the wider system. Sited and timed deliberately, with flexibility built in, the same load can instead anchor new capacity that expands access and lowers costs for everyone. Moving from the first outcome to the second is a coordination problem that no single participant owns: the grid operator, the utility, the system planner, consumer advocates, and the offtaker each hold part of it, and none can solve it alone. A financial institution can work with them to shape a configuration that meets the data center’s demand and improves the system, and to establish what it would take to finance. Solved once, a configuration like that becomes a template others can run.
The same pattern holds beyond data centers. Building retrofits make sense at scale but need aggregation and coordination before they are financeable. Industrial decarbonization is compelling in specific contexts but blocked by uncertainties no party to the transaction can resolve: whether buyers will commit over the life of the plant, and whether the energy and inputs the process depends on will be available at a cost that makes the output competitive. A regional energy transition, such as an integrated ASEAN grid, is financeable only with a regional plan, a governance structure, and a view on where assets sit. Financial institutions cannot resolve any of these on their own, but they can inform how each would need to be designed to be financeable.
No actor currently has the mandate to bring the parties who shape a system into the same room to work out what is possible, so these conversations rarely happen. In this context, it can be particularly helpful for financial institutions to say, publicly and specifically, what is blocking a given system from being financeable and what it would take to change that. What distinguishes this type of engagement from financial institutions’ routine engagement with policy makers is what is on the table: a diagnosis the other system actors can contest, rather than an ask for treatment favourable to the institution making it. That such engagement may also serve the institution’s commercial interest is not a defect; the purpose of the engagement is to achieve an outcome that aligns broader system interests with what a financial institution is able to finance.
Contributing this way is far less burdensome than the disclosure and reporting apparatus that has absorbed the sector’s effort to little effect. The benefit is first sight of a pipeline others may not yet see, and standing as the institution that can make this type of beneficial deal financeable. This is already happening in pockets: leading institutions are being drawn into exactly these system-level conversations, on data-center siting, on regional grid planning, and finding that the work is welcome precisely because no one else is doing it.
Engagement with portfolio companies should work the same way. The useful conversation is not whether a company has a target, but how it is positioned within the systems it depends on, and where it could improve that position in coordination with its suppliers, financiers, and peers, in ways that show up in margins, input costs, and ultimately valuation. The gains are the ordinary ones a company already pursues: new markets, lower input costs, resilience to disruption. When engagement is about those, it is compelling; when it is about a target set apart from what drives cost and value, it is neither realistic nor effective. A company that optimizes its position this way creates value for itself and its investors while benefitting the wider system.
None of this takes a large apparatus. It suits lean, creative teams that carry the institution’s deal-making expertise into outside conversations and connect knowledge across the firm, and it is work that internal teams tend to take up readily, because it looks for value to create rather than obligations to impose.
Redefining financial-sector climate leadership
Identifying, structuring, and enabling financeability all sit within financial institutions’ existing mandates; none assumes better risk assessment will redirect capital toward mitigation. It applies to finance the same shift our first paper made for corporates. Prevailing frameworks ask institutions to measure and disclose alignment with a modeled pathway, then treat that measurement as the contribution itself. The question instead is what makes systems financeable. Just as a firm’s climate action should be judged by how it positions itself within decarbonized systems and whether it helps transform them, a financial institution should be judged by whether it makes financeable the systems that were not. Alignment with a pathway it cannot move is the wrong test.
None of this resolves into a universal metric. An institution can only point to what it has actually done: anchoring demand that brought a first-of-a-kind project to scale, structuring a guarantee that moved a project, pressing a regulator to let project-level credit stand against the sovereign ceiling, financing an integrated system that could not be financed before. What an institution can show, not what it can report, is the measure.
