This essay was edited and structured with the assistance of Claude, an AI writing tool. The research, analysis, and arguments are my own.
Between 2016 and 2018, the brown marmorated stink bug (BMSB to its friends and foes alike) destroyed one-third of Georgia's hazelnut harvest — $60 million in damages, with some areas experiencing up to 90% crop loss. USAID's response didn't follow traditional aid patterns. Rather than providing subsidized inputs or direct assistance, the Agency facilitated a market connection between Georgian hazelnut farmers and Trece Inc., an Oklahoma-based manufacturer of pheromone traps - an environmentally friendly solution to the pest problem that fit within the broader paradigm of international hazelnut buyers looking for sustainable sourcing. Georgian farmers gained access to proven pest management technology. Trece gained access to a new international market covering 60,000 hectares across 500 villages. Both sides achieved measurable economic benefit. Parts of that relationship hardened into something self-sustaining. Other parts depended on a facilitator that no longer exists.
That distinction matters because when I and my team designed what would become the Agricultural Trade Diversification Program, we were betting on the hazelnut pattern. I left in mid-2023, and there were contracting challenges, but in May 2024, USAID/Caucasus awarded CNFA $28.975 million to run it — the successor to G-HIP (2015-2023), the vehicle through which Trece's pheromone traps reached Georgian smallholders, and USAID/Georgia's wider agriculture program. The design logic was specific: take the bilateral-economic-value pattern that had worked for hazelnuts and apply it across a broader slice of Georgia's agricultural export base. Less donor dependency per unit of intervention. More private-sector mobilization. Higher sustainability and traceability due to buyer requirements. Build the next decade of Georgian-US agricultural trade on the structural lessons the hazelnut intervention had proven.
The Administration dismantled USAID within a year of that award. The program design, the one meant to extend the architecture, was one of the casualties. The colleagues who helped stand up the hazelnut farmers' association, who brokered the Trece connection, who structured the original G-HIP Global Development Alliance — they were all RIF'd along with me.
The case matters more now precisely because the bilateral architecture that ran it no longer exists and the architecture that was meant to carry it forward got killed in the crib. The Georgia hazelnut intervention succeeded not because USAID existed in perpetuity, but because it identified where private sector incentives could align with development outcomes. That structural insight has to travel now — through private philanthropy, regional development banks, industry associations, commodity buyers with direct commercial interest — because nobody is going to run the same play with the same machinery again.
The Constraint Was Access to Technology
Georgian hazelnut farmers faced a pest management crisis that threatened their primary income source. The brown marmorated stink bug was devastating crops, but effective solutions existed, just not in Georgia. Trece had been manufacturing pheromone traps since 1984, with documented effectiveness against BMSB in US orchards where the pest was causing $37 million in apple crop losses annually and threatening $23 billion worth of susceptible crops. The technology worked and was commercially available.
The constraint was straightforward: Georgian farmers didn't know these solutions existed and couldn't access international suppliers. USAID's intervention connected Georgian farmers with Trece, established distribution channels through farmer associations, and funded capacity-building programs with Penn State and USDA to ensure farmers could use the technology effectively.
This approach differs fundamentally from traditional aid models that provide subsidized inputs or implement solutions directly. USAID facilitated a market connection between parties who both benefited economically. The relationship sustained itself commercially after USAID program funding ended because both parties had ongoing economic incentives to maintain it.
Bilateral Economic Value as Sustainability Model
The Georgia-Trece partnership demonstrates that well-structured interventions create bilateral economic value rather than one-way resource transfers. Trece gained access to international markets serving 500 villages across 60,000 hectares. Georgian farmers gained access to proven pest management technology. By 2024, Georgian hazelnut exports recovered to $90 million (15,000 tons) with prices increasing 16-17% to over $6/kg. Ferrero invested directly in Georgian farms after establishing stable supply. American agricultural researchers gained field validation for BMSB management science.
Which of those relationships actually survived USAID's destruction in 2025 is a more honest and more interesting question than the answer the field is giving itself. Ferrero's relationship with Georgian producers seems to have hardened — 88% of Georgia's in-shell hazelnut exports went to Italy in 2022, overwhelmingly through Ferrero's supply chain, and that buying relationship is still active for now. Whether it remains, with the loss of USAID and Georgia's democratic backsliding remains to be seen. A buyer with enduring commercial interest in a specific origin shouldn't need a bilateral agency to keep that relationship alive. It should self-fund.
But Trece's reach into Georgian smallholders was always mediated through USAID-funded CNFA programming — the programmatic procurement channel, the training infrastructure, the demonstration plots. No public evidence I can find in 2026 shows Trece with direct commercial distribution into Georgian farms outside that channel. It may have transferred to Ferrero's sustainability programming. It may have quietly reverted to pre-intervention levels. It may persist through private imports at smaller scale. I don't know, and I suspect nobody outside a specific commercial review has a clean answer.
That ambiguity is not a flaw in the argument. It is the argument. Well-designed bilateral interventions create relationships that fall along a spectrum. On one end are the ones that harden into commercial self-sufficiency — Ferrero-Georgia is that. On the other end are the ones that need an ongoing intermediary to keep the connection alive, and those relationships last only as long as the intermediary does. Trece-Georgia appears to have sat closer to the second end than the first, and that is truly unfortunate, both for the hazelnut farmers and Trece themselves.
The design discipline isn't about guaranteeing every relationship hardens. It's about being honest about which ones will and which ones won't, and designing accordingly — either engineering genuine commercial alignment, or naming the ongoing facilitation cost before the intermediary disappears. USAID's dismantlement is not just a stress test, a chance to gauge performance, but a pulling of the rug well before such a test would have been reasonable.
What Works When Donor Institutions Are Unreliable
The destruction of USAID and the diminished state of bilateral development instruments creates strategic imperatives for practitioners, funders, and private sector actors who want the work to continue. The Georgia case offers three applicable frameworks:
1. Align Private Sector Profit With Development Outcomes
The hazelnut-Trece connection worked because Trece's business case — selling pheromone traps — aligned with development goals — protecting farmer livelihoods. The better their product worked, the more farmers bought, the more Trece made and could invest in R&D for new products. No perpetual subsidy required.
Map development challenges against existing private sector technologies. Where companies have developed solutions for domestic markets, identify parallel problems in developing countries where those solutions could create commercial value while solving development challenges.
2. Build Market Infrastructure, Not Project Dependencies
USAID's G-HIP Global Development Alliance strengthened the Georgian Hazelnut Growers' Association, established 8 processing facilities, and created a network for agricultural input suppliers. That infrastructure made it viable for companies like Trece to serve the market commercially. Farmer associations and retailers handled distribution. Processing facilities created consistent demand, and the higher quality demanded by valuable (non-Russian) markets.
Invest in market systems — industry associations, processing infrastructure, training capacity — rather than direct aid programs. Assess success by whether private sector actors can operate profitably within the strengthened system.
3. Facilitate Connections Before Crises Emerge
USAID didn't develop technology or distribute traps directly. They connected Georgian farmers with Trece, funded capacity building, and established distribution channels. Then they stepped back. The constraint was access to existing, proven technology — pheromone traps were already solving the same pest problem in US agriculture.
Map existing private sector technologies solving problems in developed markets against parallel challenges elsewhere. Foundations, private philanthropy, or industry associations can facilitate these connections. Invest in the connective tissue — market research, relationship building, capacity building — that allows commercial relationships to form.
The Lessons That Outlived the Agency — and the Program That Didn't
USAID ran the Georgia hazelnut intervention. My team and I designed what was supposed to carry it forward. The Administration destroyed USAID before the successor could prove itself. All three things are true, and the case keeps teaching.
It teaches that bilateral aid, when it works, builds infrastructure that can stand without it — and that the design discipline for doing that is knowable, transmissible, and worth defending. Ferrero, I believe, still buys Georgian hazelnuts. The farmers' associations still function. The field-ready BMSB management science that doesn't require perpetual Agency support persists. Those pieces hardened. The Administration's dismantlement of USAID didn't erase what the commercial parties had absorbed into their own operations.
But not every piece of what we built was supposed to run on its own yet. The Agricultural Trade Diversification Program was designed to do the intermediary work for another five years — specifically because we knew some relationships needed more time to harden, and some market-making was still too early to self-sustain. The potential we lost, from the purely farmer production side to the countering Russian market control geopolitics, is an opportunity lost forever. Thankfully, though, the knowledge of how it was designed, what it was meant to catalyze, which commodity systems were poised to transition to commercial self-sufficiency given one more round of facilitation — that knowledge did not die with USAID. It lives in everyone who designed, implemented, and evaluated that work.
This is the rebel-bureaucrat move. Not nostalgia for an agency that is gone. Not grief for a program that never ran. Craft, intact, waiting to be picked up by whoever is willing to do the next version of this work — private foundations, regional development banks, commodity buyers with direct interest, industry associations, local governments. The Georgia case is a blueprint. The ATDP design is a more evolved one, and it is still available to anyone who asks.
Where do development challenges in your sector align with existing private sector capabilities in ways that could create sustained and sustainable bilateral economic value? And when you design the intervention, are you being honest with yourself about which relationships will actually harden without an ongoing facilitator? Those are the questions the Georgia case keeps pointing me back to.