CREATING WITH SYSTEMS

STRUCTURAL RESILIENCE
IN A LOCAL FOOD SYSTEM.
Why King County's Farm to Food Bank Programs are Resilient by Design.

// 01 IDENTIFYING THE ARCHETYPE

IDENTIFYING
THE ARCHETYPE

Like many counties in the US with a strong local food system, King County has a variety of farm-to-food bank programs. Within these programs is a strategy called value-based procurement. The pattern is straightforward: a funding source provides hunger relief organizations (HROs) with funds, which they use to buy food directly from local farms through term-based contracts. The key is that the funding enables contracts to be profitable for farmers, making local farming more viable, while simultaneously providing a stable, flexible, and healthy food supply to food banks, which boosts food security. Additionally, the money these funds provide stays in the region and compounds. It is a clever, systemic way to get the most value from funding.

In the US, the model began emerging at a federal level as early as 1938 and evolved iteratively into various programs, most notably The Emergency Food Assistance Program (TEFAP). Regional incarnations have also recurred independently, such as Pennsylvania's PASS program, Vermont's Vermonters Feeding Vermonters, and New York's Nourish New York.

As with all such systemic patterns, the proof is not in the theory but in the implementation. The basic model, as described above, can be fragile and prone to shock either from funding sources or struggle with the level of coordination required to make the pattern functional. Such fragility is a significant flaw when the purpose is to make long-cycle endeavors, such as farming and critical food infrastructure, more secure and reliable. This system's quality is most apparent when it is put under stress.

An example of such a shock was experienced by all such programs across the country when the federal funding that underpinned many instances of the model were aggressively and unexpectedly cut in 2025. Washington State alone lost roughly $25 million worth of farm-to-food bank funding overnight.^[1] This was not a natural response to a lack of demand from hunger relief organizations. Food Lifeline serves 1.7 million people across Western Washington, which is double the pre-pandemic number, and statewide food bank visits hit 13.4 million in 2024.^[2]

However, while the cuts certainly affected King County, many procurement-based farm-to-food programs survived and even grew. So, what is happening in King County that makes it so resilient?

The overall farm-to-food bank system in King County is comprised of a network of organizations, each with distinct roles in addressing food security and local agriculture. Some organizations specialize in targeted interventions or serve particular segments of the food system, while others connect these specialized initiatives into a larger, coordinated framework. To understand the region's unique resilience, we will focus on one organization that has played a significant role in actively curating, collaborating, and orchestrating values-based procurement within the diverse set of actors in this network: Harvest Against Hunger (HAH).

A quick note before we proceed: Writing about complex systems often constrains the narrative to tidy cross-sections extracted from what is a highly enmeshed reality. This post is no exception, yet while creating a cross-section is not comprehensive, it can be very useful; much in the same way understanding a respiratory system is useful, yet it represents neither the full creature or an independent creature in itself. With that said, let's proceed with the central focus being how HAH's strategic partnerships related to farm-to-food bank procurement work, and examine the specific functional structures that underpin its resilience.

Founded in 1984, HAH serves as a coordination layer, connecting growers, markets, logistics providers, and hunger relief organizations to form an integrated system. Across all its initiatives, the organization delivered over 5 million healthy servings of produce in fiscal year 2023–2024, representing approximately 586,636 pounds.^[3][12] HAH is highly collaborative and facilitative, actively developing relationships and managing programs across a wide range of players, including collaborations, for example, with Neighborhood Farmers Markets, Cascadia Produce, and the Good Business Network of Washington.^[30][31][32]

When the cuts hit, the loss was devastating to many hunger relief programs. However, HAH's own managed programs (many of them collaborations with other organizations) continued to operate, grow, and allocate resources to shock-affected parts of the local system. One program expanded farm participation by over 25%,^[4] and when SNAP and WIC disruptions compounded the pressure on food banks, HAH assembled a new coalition called Cooperating to Feed Community, bringing private partners together to fund food purchases directly for eight hunger relief organizations across King County.^[27]

This is not the first time HAH has facilitated the food system growing stronger during a crisis. When COVID-19 shut down Seattle's farmers' markets in 2020, small farm vendors lost their primary sales outlet overnight. HAH, in partnership with NFM and PCC, responded by creating Growing for Good, an emergency program that channeled private funding into farm-to-food-bank contracts to alleviate the missing income. The program was so successful that it never ended and became a permanent part of the farm-to-food bank infrastructure.

In short, procurement strategies are alive and well in King County even after major external shocks, and HAH has an orchestrating role in all of them. Any system that repeatedly builds new capacity out of successive crises is impressive, so let's discover how it works.

// 02 FUNCTIONAL STRUCTURES

WHY IT
WORKS HERE

Systemic design often entails looking for patterns that are known to exist for a given set of behavioral outcomes. A software engineer or a biologist might go looking for underlying patterns and sub-layers to understand machine or natural behavior, and a systemic designer will do the same for cross-domain behavior. Here are the main functional structures employed by HAH that give it such a central role in the resilience we are interested in (in no particular order):

• Facilitated Decentralization
• Diversification & Redundancy
• Lifecycle Design
• Delay Management
• Interfaces

Let's jump in.

Facilitated Decentralization:

Facilitated decentralization is when one actor performs strategic work to connect and establish relationships, while leaving the network decentralized. HAH does not own farms, warehouse food, or directly operate a fleet of trucks, but it does bring all these resources together from independent actors. The participants make their own decisions, but the facilitator makes it easier for those decisions to connect and emerge as a functional whole.

This pattern has many benefits: it can field capacities that would be prohibitively expensive to build/own directly. It is flexible, highly adaptable, and it leverages system actors' own motivations to form an incentives-based, common outcome. What this strategy lacks in the speed and resource efficiency often associated with centralized topologies, it makes up for in the previously identified benefits, which are quite appropriate to a local food system.

Diversification / Redundancy:

Redundancy in systems creates resilience by preventing single-point or cascading failure. Relying on three distinct funding sources to power the farm-to-food-bank procurement funds, rather than just one, isolates the overall system from single points of failure, a key component to resilience. Diversification is key; because the redundancy is diverse, it is not wastefully duplicative, but a deliberate arrangement of contingencies.

The three programs each draw from entirely separate funding wells that can be directed to support farm-to-food bank.

King County Farmers Share (KCFS) is funded by King Conservation District, a local conservation agency whose assessment rate was renewed through 2029.^[11] KCFS was built specifically for beginning farmers new to wholesale, offering both purchasing contracts and technical assistance to help them build capacity for institutional sales. Now in its sixth year, it works with over 40 small farms and 30 hunger relief agencies.^[4][9] It is also actively rebuilding parts of the system damaged by the 2025 federal cuts, with a particular focus on new and underserved farmers.

Growing for Good (GfG) represents something unusual and remarkable: a fully private funding source. It is funded entirely through PCC Community Markets shopper round-up donations, making it one of the few programs of its kind that is not dependent on public money.^[5]

Farm to Food Pantry (F2FP) is the oldest and most scaled of the three. It is funded by the Washington State Department of Agriculture (WSDA) through state legislative appropriations and grew out of a simple insight from grower roundtables in 2014.^[6] F2FP now operates through 45 agencies across 34 counties and is active through June 2027.^[8]

The three programs are funded from separate, independently governed sources. When federal funding for programs like LFPA and TEFAP vanished in 2025, systems built only on those sources had nowhere to turn. Because HAH's programs are funded at the regional level (and even then at the county, private, and state levels within), it can withstand shocks and preserve local systems. That redundancy is a product of the diversified structure, but it also comes with another interesting benefit.

Lifecycle Design:

The three programs are not just diversified by funding source. They also serve farmers at different stages of development. KCFS is for new farmers. It pairs purchasing contracts with technical assistance so that farmers who have never sold at an institutional scale can learn how to do so. Growing for Good sits in the middle, as its farms have experience selling goods at farmers' markets and are used to delivering to meet demand. F2FP operates at the state level across 34 counties and can support a wide range of procurement relationships, from smaller farms who meet the procurement criteria to producers already working at scale.

Not only does this mean that the three programs are already positioned to meet farmers where they are, but farmers can also grow and scale without phasing out of the relationships with HROs. The structure supports scaling in both directions: as farmers expand their capacity and move from local pilot contracts with KCFS to broader engagement through Growing for Good or F2FP, the system allows for gradual adaptation rather than requiring a sudden leap or cutoff. This adaptive approach enables both farmers and HROs to participate at levels that suit their current stage, while also providing clear pathways for growth and deeper integration in the network.

Delay Management:

Growing food usually means spending money on expensive, high-risk operation with months of risk exposure before earning any return. In a traditional reimbursement model, a farmer effectively finances the program out of pocket or on a loan until the check for the produce arrives. For a small or beginning farmer, that gap between spending and getting paid back can be the difference between participating and not.^[33][34]

Growing for Good addresses this directly. Rather than reimbursing after delivery, GfG provides pre-season capital and upfront spring payments so farmers are not exposed to financial risk during the most vulnerable part of the year.

This is an excellent example of how delays can create significant barriers within systems, an often underestimated or overlooked element to a successful structure.

Interfaces:

Interfaces are points of discovery, communication, and expectation. The quality of these connection points often determines how well the overall system performs, as poorly designed interfaces create friction, misunderstandings, and chaotic echoes that compound across the whole network. A relationship-focused organization like HAH is automatically "in the business" of interfaces.

A great example of how HAH uses interfaces is how it matches farms to HROs. Discovering the three programs, funding sources, and respective sets of requirements are a lot for a farmer to understand and navigate. So, HAH provides a single intake form, the Match-Up Survey, that reduces the need for farmers to understand the internal workings of various funding programs and the criteria for starting a relationship with an HRO. A farmer describes their situation once, and HAH recommends the best procurement partner based on the specifics of the procurement programs. This matching is not an automatic assignment, and the relationship is ultimately established by the HRO with the farmer; however, this streamlined matching is a valuable component of orchestration.

Consider the alternative: three separate programs to discover, contact, understand, and negotiate with for the farmer must take time and effort to decide which to pursue. That is a lot to ask from one of the most time-intensive professions to exist. That kind of complexity can easily keep small and beginning farmers from entering wholesale markets in the first place. By absorbing that complexity into a single interface, HAH lowers the barrier to entry for the farmers who need it most.

These aren't coincidental features, rather, they are the result of 40 years of institutional learning and hard work by the organizers. Each of these highlights the structural qualities that contribute to the ecosystem's demonstrated resilience.

// 03 PARTICIPATION

HOW CAN YOU
PARTICIPATE/SUPPORT?

We all want to invest in working systems, so how can we support and participate in this one?

Start by talking about it:

King County residents may not know that a 40-year-old nonprofit coordinates three programs that connect local farms to food banks. Especially in an age when institutional trust is in very short supply, discussing with peers that an organization achieves real outcomes is a low-cost way to increase support for a working system.

• HAH: Match-Up Survey / King County Farmers Share
• HAH: Programs and Impact

Shop within it:

Small shifts in purchasing behavior are powerful. It takes a bit of effort to change a purchasing habit to support local food systems, but if you can form the habit, you are consistently participating in a meaningful way. Use the round-up option to contribute to GfG if you shop at a PCC market. Additionally, at Neighborhood Farmers Markets, buying from GfG partner farms puts money into the same network. If you use SNAP/EBT, Market Match doubles your purchasing power at seven NFM locations. It is important to make a plan that builds a habit: make a trip to the farmers market a regular social habit, and choose something to buy that puts you in front of the PCC round-up. Most people buy food often, and bundling that necessary behavior with support might be the best way for many to meaningfully engage in a stronger region.

• Growing for Good
• GfG partner farms at Neighborhood Farmers Markets
• NFM: Food Access programs (SNAP, Market Match, Fresh Bucks)

Volunteer within the ecosystem:

Volunteering is another way to participate directly. There are opportunities at multiple levels of the system.

The most direct is volunteering at one of the hunger relief organizations in HAH's network. The 23 organizations that receive food through Growing for Good include Ballard Food Bank, Rainier Valley Food Bank, University District Food Bank, West Seattle Food Bank, and others across the county. KCFS additionally works through the Seattle Food Committee, Hopelink, South King County Food Coalition, and Mary's Place. These are the endpoints of the system described above, and they always need hands.

For those drawn to the production side, HAH manages the Washington Gleaning Network, which organizes collaborative harvests that recover food that would otherwise go unharvested. Lettuce Link, a program of Solid Ground, runs a similar hyperlocal gleaning operation through Seattle's P-Patch community gardens, with volunteer delivery three days a week during the summer.

Some local farms also welcome volunteers directly. Marra Farm in South Park (Solid Ground) produces organic vegetables, which are donated to food banks, with drop-in hours on Wednesday, Friday, and Saturday. Tilth Alliance runs volunteer programs at Rainier Beach Urban Farm, Seattle's largest urban farm.

• HAH: Volunteer Hub
• HAH: Washington Gleaning Network
• Lettuce Link / Solid Ground: Helping to Harvest
• Marra Farm: Giving Gardener Volunteer
• Tilth Alliance: Volunteer

Stay aware of the policies that support it:

Funding is undeniably an important input, and each of the three funding streams has its own policy backstory. King Conservation District assessment, state appropriation, and PCC round-ups. Each one exists because someone built it, advocated for it, and maintained it. Food Lifeline, Northwest Harvest, and the Anti-Hunger & Nutrition Coalition all run advocacy alert lists. Signing up takes a minute and means you'll hear when a specific bill or budget line needs constituent pressure.^[10]

• Food Lifeline: Advocacy
• Northwest Harvest: Advocate
• Anti-Hunger & Nutrition Coalition
• King County: Local Food Initiative

Make a recurring donation:

Recurring donations let organizations plan, hire, and commit to multi-year programs. HAH has a 98% rating from Charity Navigator, with 95% of expenses going to programs. A $25/month Harvest Partner commitment delivers 770 lbs of produce through F2FP. Growing for Good has its own community sponsorship. Both fund distinct parts of the system: HAH funds coordination, and GfG funds the private procurement stream.

• HAH: Harvest Partners — Monthly Giving
• HAH: Donate
• GfG: Direct Donation
• GfG: Community Sponsorship Levels
• Charity Navigator: HAH Rating (98%, 4-star)

// 04 DESIGN FRONTIERS

WHERE DOES IT
GO FROM HERE?

Systemic design is not just a tool for analysis, and in this section we will show how it can be used as a framing for creative system development. There is a rich landscape of patterns and analogs the programs we have been discussing can draw from as they pursue opportunities and face immediate challenges.

The F2FP Annual Report, conducted by HAH and WSDA, outlines numerous challenges sourced from actors within the system.^[7] What follows in this section are a few of the findings viewed through a systemic design lens. To be clear, these are not prescribed solutions, but to showcase solution framings that, given the proper research and adaptation process, could help the overall system grow.

Cold storage and logistics:

The Annual Report shows 54.8% of food banks in the program have limited cold storage, one of the most frequently cited capacity constraints. But with the right framing, the constraints of cold storage on a site might be addressed in ways beyond adding more freezers as physical infrastructure; the more fundamental question is how food moves through the system. There are at least three ways to think about movement and flow that might reduce reliance on physical cold storage: how cold storage is distributed across the network, how quickly food is matched to supply and demand, and whether it can be processed into more shelf-stable forms before it ever needs refrigeration.

Regarding how cold storage is distributed, where it sits in the network can be as important as how much there is. HAH's strength is coordinating relationships, and that capacity could extend to how cold infrastructure is shared and distributed across the network. There are models for this already operating within or adjacent to F2FP. The NEW Hunger Coalition built a refrigerated micro-delivery system using CoolPup trailers within F2FP itself, serving food banks within a 57-mile radius on data-driven, predetermined routes.^[15] The cold storage is in the trailer, which means a small food bank that could never justify a walk-in freezer still receives refrigerated food on schedule. The storage is mobile and modular, meaning it could be shared dynamically and efficiently across sites. Hopelink's Mobile Market in King County brings distribution directly to 12 rotating communities,^[16] with food arriving and being distributed in the same visit, centralizing cold capacity in the vehicle rather than requiring it at every endpoint. The pattern across these is that cold infrastructure is moved to the transport layer and serves multiple sites rather than a single one. An organization like HAH, which already coordinates relationships between farms, food banks, and logistics providers, is well-positioned to facilitate this kind of shared infrastructure, whether that means organizing grants for refrigerated trailers, brokering shared cold storage between nearby sites, or expanding the mobile distribution model so that more food reaches people directly rather than waiting in a food bank freezer.

On speed, a practice called virtual food banking is emerging where digital coordination reduces or eliminates the time food spends waiting in storage. Instead of aggregating food at a central warehouse until it is claimed, matching platforms connect available supply directly to immediate demand. Feeding America's MealConnect has rescued 6 billion pounds of food this way, and all 200 Feeding America food banks now use the platform.^[25] It recently expanded to include a Produce Marketplace where growers post surplus, and food banks order based on their own demand and capacity. Globally, virtual food banking doubled from 5% to 11% of the Global FoodBanking Network's volume between 2022 and 2023.^[14] A 2025 peer-reviewed study of 200 food banks found that just-in-time distribution strategies, where food reaches clients faster through mobile and direct delivery, significantly reduced food waste.^[26] There are already two adjacent players in the Puget Sound region using variants of this model: The Good Food Exchange and Good Roots, which pioneer digital platforms for coordinating donated food exchange^[28] and contactless food access^[29] respectively. How this principle applies to a network like F2FP's, where the food is fresh, local, and perishable, would require contextual adaptation, but the underlying idea, that faster coordination reduces the need for physical buffers, is well established.^[13]

On processing, Good Shepherd Food Bank in Maine operates a subsidiary that converts surplus blueberries and vegetables into shelf-stable products, transforming perishable food into inventory that lasts months rather than days.^[17] Closer to home, Kitsap Conservation District's Farm to Freezer program processes over 10,000 pints per year of soups and salsas from gleaned and rescued produce, keeping roughly 20,000 pounds out of landfills.^[18] If rules around local food definitions are being refined anyway, and within F2FP, they have been, with the Decision Tree formalized in SFY23 and revised in SFY25, there may be room to accommodate locally processed and preserved foods in ways that shift the cold storage constraint upstream before it ever reaches the food bank.

A freezer grant, a shared refrigerated trailer, a virtual matching platform, and a processing facility are all possible responses to the challenge. They also are not competing ideas but may be complementary, and the question for any given context is which combination makes the most sense given the density, distance, and infrastructure already available.

Funding through interaction design:

The Annual Report identifies funding as the primary challenge across the ecosystem, with 56.7% of HROs reporting that they need more funding. While F2FP itself is funded through state appropriations, the broader funding landscape that supports this system can be affected by alternative sources. The previously discussed Growing for Good, for instance, is a fully private stream built on PCC shopper round-ups, and it has become a significant part of the ecosystem. How it was designed is worth paying attention to and may offer potential mitigations for the funding gap. Research shows that 45% of shoppers agree to round up at checkout, compared with 18% who agree to a flat donation.^[19] It is the same money; however, the way donors interact with the choice is a fundamentally different design. The round-up works because it removes the cognitive burden of deciding to donate, embedding funding into something you are already doing. This draws on principles from behavioral and interaction design, focusing on aligning with behaviors people already use rather than asking them to adopt new ones. That thinking can be applied here: what integrated events or experiences could be developed more effectively to remove barriers from willing donors/participants?

The checkout charity industry as a whole raised $749 million in 2022, a 24% increase from 2020, with 67% of campaigns now offering round-up options.^[20] The question is where else this principle can apply. A farmers market visit, a CSA signup, and a restaurant meal each represent a moment where a well-designed, low-friction, ethical ask could generate a new independent channel.

Some such models already exist. In Pittsburgh, 412 Food Rescue's FarmShare program lets CSA subscribers buy one share and sponsor one share, delivering 4,675 pounds of food in 2024.^[21] In Durham, Farmer Foodshare's "CSA for All" uses a sliding-scale subscription where everyone gets the same box.^[22] In Massachusetts, Waltham Fields' "Food for All" program adds a $100 option that subsidizes half-price shares while dedicating 20% of the harvest to food banks.^[23]

While these are based on two of many challenges reported by F2FP, the specific design patterns and examples are less important than the process itself. Reframing cold storage as a question of how food moves through the system, or reframing funding as a question of how giving is designed into existing behavior, are ways of looking at known challenges that surface new possibilities. That process of looking structurally, researching analogs, and adapting what fits is the same process that built this system to this point and a glimpse of how it may develop.

// 05 CONCLUSION

A SYSTEM WORTH
INVESTING IN

King County's farm-to-food bank system, thanks in part to Harvest Against Hunger's excellent contributions, is a working example of what local food infrastructure looks like when it is built well. The archetype is not unique to King County, but the implementation is. Years of institutional building produced a system where funding is diversified across county, private, and state sources, where farmers can enter at their level of readiness and grow through the programs, where the timing of capital reduces a growers financial exposure, and where a single interface absorbs the complexity that would otherwise overwhelm participants. When federal funding disappeared or a pandemic shut down markets, this system kept operating and growing.

You can support it directly by shopping within it, staying aware of the policies that fund it, making a recurring donation, volunteering, or simply talking about it with others.

In an age where we need working systems, here is one worth maintaining and supporting.

Special thanks to Olivia Jackiewicz and Mary Embleton, who helped me validate my finding and more accurately depict the systems they have built and continue to shape.