The $1 Million New Business Trigger: USDA B&I Feasibility Rules in Plain English

Every few weeks a lender or a borrower sends me a link to some page that explains when a USDA Business and Industry loan needs a feasibility study, and asks whether it is right. Most of the time it is not. Not wildly wrong, usually, but wrong in the places that matter: the page says "over $1 million" and stops, or it says "every startup," or it cites a part of the Code of Federal Regulations that has not governed B&I loans since 2020. One page I looked at last month tells readers that a study is required for "any new business, expansion, or acquisition." That is a fine way to sell feasibility studies. It is not what the regulation says.

I sell feasibility studies too, so take what follows with that in mind. But I would rather people hire us because they understood the rule and decided they need the work, not because they were frightened into it. So this is the rule, in plain English, with the section numbers attached so you can check me.

The short answer

There are three tiers, not one threshold. This is the thing almost nobody gets right.

Tier one: guaranteed loan of $600,000 or less. Your lender submits the streamlined application package under 7 CFR 5001.306(b). There is no automatic feasibility study. USDA can still ask for one, but it has to have a reason (1).

Tier two: guaranteed loan above $600,000, but either $1,000,000 or less, or the borrower is an existing business.The full application package under 5001.306(a) applies. A feasibility study is discretionary. USDA "may require" one when the lender's analysis, the business plan, or the project information is not enough to establish that the project is technically feasible and economically viable, or when the project would significantly change how an existing business has historically produced cash (1).

Tier three: guaranteed loan greater than $1,000,000 to a new business. A feasibility study prepared by an independent qualified consultant acceptable to USDA is required. Full stop. There is no waiver mechanism in the text. The only thing USDA controls is the scope (1).

That is the whole rule. Everything else in this article is about what the words in it actually mean, because "guaranteed loan," "new business," and "independent qualified consultant" are all defined terms, and each definition has a trap in it.

What the regulation actually says

The B&I program used to live at 7 CFR Part 4279. On October 1, 2020, USDA folded B&I, Community Facilities, Water and Waste Disposal, and the Rural Energy for America Program into a single guaranteed-loan regulation, 7 CFR Part 5001, known as OneRD. The final rule was published at 85 FR 42494 on July 14, 2020 (2). If a page you are reading cites Part 4279 or Part 4280 as the governing B&I rule, it is out of date or was written by someone who did not check.

The operative language for B&I is 7 CFR 5001.306(a)(3). I am going to quote it because the exact wording matters:

"The Agency may require a feasibility study when the lender's analysis, borrower's business plan, or project information is not sufficient to determine the technical feasibility, market feasibility, or economic viability of the project.(i) For guaranteed loans greater than $1,000,000.00 to a new business, a feasibility study prepared by an independent qualified consultant acceptable to the Agency is required. The scope of the feasibility study will be determined by the Agency and is dependent on the complexity of the project and the borrower.(ii) For loans of $1,000,000.00 or less to new and existing businesses, the Agency may require a feasibility study when the lender's analysis or other borrower information is not sufficient to determine the technical feasibility or economic viability of the project, or if the project will significantly affect the operations of a borrower who is an existing business and its historic cash flow." (1)

Notice the structure. The opening sentence is a general grant of discretion. Paragraph (i) is the one mandatory case. Paragraph (ii) is discretion again, spelled out for the sub-$1 million band. There is also a residual authority buried in the general application section, 5001.303(c)(4), which says that if USDA cannot find "a basis for successful repayment" from the lender's analysis, the business plan, or other project information, or if the project will have significant impacts on existing operations, it "may require an independent feasibility study" (3). That residual clause has no dollar figure in it at all. It is how USDA ends up asking for studies on $700,000 deals when the lender's package is thin.

So the honest summary is: above $1 million to a new business, the study is mandatory by rule. Everywhere else, it is a judgment call, and in the current environment USDA is making that call more often than it used to.

"Guaranteed loan" means the loan, not the project

The first definitional trap. The $1,000,000 figure is measured on the guaranteed loan amount. Section 5001.306 opens by splitting applications according to whether "the guaranteed loan amount is more than $600,000" or not, and paragraph (a)(3)(i) refers to "guaranteed loans greater than $1,000,000.00" (1). Under the definitions in 5001.3, a guaranteed loan is the loan on which USDA issues a loan note guarantee (4).

Three things that number is not:

It is not the total project cost. A $6 million hotel financed with $2.5 million of equity, a $2.6 million conventional first mortgage, and a $900,000 B&I guaranteed loan does not trip the mandatory trigger, because the guaranteed loan is $900,000. USDA can still ask for a study under its discretionary authority, and for a ground-up hotel it very likely will, but the rule-based requirement is keyed to the guaranteed loan.

It is not the "guaranteed portion." This is where sloppy summaries go wrong. A guaranteed loan has a guaranteed portion (the loan amount times the guarantee percentage, which for FY2026 is 85 percent under $5 million and 80 percent at or above it) and an unguaranteed portion (5). The trigger uses the defined term "guaranteed loan," meaning the whole loan the lender makes under the OneRD guarantee, not the 85 percent slice USDA stands behind. A $1.1 million guaranteed loan with a $935,000 guaranteed portion is over the threshold.

It is not something the regulation explicitly aggregates for the study trigger. The B&I borrower ceiling of $25 million at 5001.406(c) is expressly aggregated across guaranteed and unguaranteed portions, existing balances, and new applications (6). The feasibility trigger in 5001.306(a)(3) is written per application and contains no such aggregation language. I flag this as an ambiguity rather than a loophole, because a borrower who tries to split a $1.8 million project into two $900,000 loans to dodge the study will run straight into 5001.303(c)(4), and USDA will ask for the study anyway. Phased projects have their own rule: each phase has to be financially sustainable on its own, without counting on future phases.

"New business" is a term of art, and it is broader than you think

The second trap, and the one that generates the most arguments.

Here is the definition from 7 CFR 5001.3, in full:

"New business means a business that has been in operation for less than one full year and a business that has been in operation for at least one full year and has not achieved full operational capacity or stable operations as determined by the Administrator, including a new enterprise or new affiliate of an existing business moving or expanding into a new location involving new market or labor areas." (4)

And its mirror image:

"Existing business means a business that has been in operation for at least one full year and has achieved full operational capacity or stable operations as determined by the Administrator. The following will be treated as existing businesses provided there is not a significant change in operations of the existing business: Mergers by an existing business with a new or existing businesses, a change in the business name, or a new business and an existing business applying as co-borrowers." (4)

Read the first definition twice. It has two prongs and a rider.

Prong one: under a year of operations. Simple. If the doors opened ten months ago, the borrower is new.

Prong two: over a year but not stabilized. This is the prong that surprises people. A business that has been running for fourteen months but is still ramping toward capacity, still losing money, or still has volatile month-to-month results is a new business "as determined by the Administrator." There is no numeric test for "full operational capacity or stable operations." The determination is made, in practice, by the Rural Development State Office reviewing the file. I have seen a two-year-old business classified as new because its third year projections bore no resemblance to its first two.

The rider: new affiliates and new locations. A new enterprise or a new affiliate of an existing business that is "moving or expanding into a new location involving new market or labor areas" is a new business by definition. A regional trucking company with fifteen years of history that forms a subsidiary to open a truck stop three states away has created a new business, regardless of how seasoned the parent is.

Some of the edge cases that come up constantly:

A newly formed entity acquiring an operating business. This is the single most common misunderstanding. The borrower is the new entity. It has no operating history. USDA will generally treat it as a new business, even though the café or the car wash it is buying has been open for a decade. Whether the State Office is willing to look through to the underlying operation and call it stabilized is a case-by-case judgment, and the safe assumption above $1 million is that the study is required.

Buying a closed business or converting a property to a new use. No historical debt service coverage for the proposed operation. Expect a study.

A startup franchise. New business. The franchisor's pro forma is not a substitute; the study has to test demand independently.

Expansion of an existing firm into a new product line or a new market. If it runs through a new affiliate or a new location in a new market or labor area, it is inside the definition. If it is the same entity in the same market adding capacity, it is probably an existing business, but if the expansion will "significantly affect" the historic cash flow, USDA has the discretionary hook under (a)(3)(ii).

A pure change of ownership of a stable business, same entity, no operational change. Generally existing. A name change or a merger with no significant change in operations is expressly treated as existing (4).

Refinancing. Not a new business question as such, but where the refinance lacks historical coverage for the new debt, USDA can and does reach for 5001.303(c)(4).

The practical advice is boring and correct: if the borrower is anywhere near the line, get the State Office to say in writing, before you spend money on a study, whether they consider the borrower new or existing. That letter is worth more than any consultant's opinion, including mine.

Can the study be waived above $1 million? No.

I get asked this every month, usually by a lender with a strong, well-collateralized deal and a borrower who does not want to spend fifteen thousand dollars. The answer is that "waiver" is not a concept the regulation offers for tier three. Paragraph (a)(3)(i) says "is required." What USDA can adjust is the scope: "The scope of the feasibility study will be determined by the Agency and is dependent on the complexity of the project and the borrower" (1). A simple single-revenue-line project might get a lighter scope than a biorefinery. It will not get no study.

The only real path out of a mandatory study above $1 million is to establish that the borrower is an existing business, and that is a classification question, not a waiver.

Below $1 million and for existing businesses, the calculus is different. USDA can decide a study is unnecessary because the lender's analysis and the business plan already establish feasibility. That decision belongs to USDA, and the way to earn it is a genuinely thorough lender package, not a request.

Who is an "independent qualified consultant"

Third trap. The regulation does not maintain a list of approved consultants. There is no roster, no certification, no "USDA approved" status, despite what some marketing pages imply. Acceptability is decided by the State Office reviewing the file (7).

What the regulation does define is the phrase itself. Under 5001.3, a qualified consultant is "an independent third-party person possessing the knowledge, expertise, and experience to perform the specific task required" (4). Two elements: independence, and expertise matched to the task. A hospitality analyst with twenty years of hotel work is a qualified consultant for a hotel. The same person is not obviously qualified for a meat processing plant, and a State Office is entitled to say so.

Independence is enforced through the conflict of interest definition, also in 5001.3. A conflict of interest is "a situation in which a person has personal, professional, or financial interests that prevent, or appears to prevent the person from acting impartially," and the definition expressly captures a person acting as a compensated agent of both the borrower and the lender on the same loan, distributions of loan funds to owners and insiders, and refinancing debt owned by a loan packager, broker, or referral agent or its affiliates (4). The enforcement mechanism at 5001.208 is short: "The lender must report all conflicts of interests, in writing, to the Agency" (8).

Some consultant sites describe 5001.208 as an elaborate independence code. It is one sentence. The substance is in the definitions. But the practical effect is clear enough:

• The lender cannot write the study for a loan it wants USDA to guarantee.

• The loan packager cannot write the study for the deal it packaged.

• Anyone paid a success fee contingent on approval fails the "appears to prevent" test.

• The borrower's regular accountant, whose relationship is not arm's length, is a bad choice and a State Office can reject it.

• The income approach of an appraisal is not a feasibility study. That line comes from the legacy rule at 7 CFR 4279.150 and USDA has carried the principle forward in practice (9).

  
  

For what it is worth, our own rule at MMCG is stricter than the regulation: we will not prepare a study on a deal we have packaged, brokered, or hold any ownership or contingent-fee interest in, and our fee is split fifty-fifty and paid regardless of whether the loan closes. That is not a USDA rule. It is a reflection of the fact that a study's only value is its independence, and a study whose author gets paid more if the loan closes is worth exactly nothing to a credit committee.

What the study has to contain: five components, thirty-seven factors

The 5001.3 definition of a feasibility study fixes the content. It is "a report including an opinion or finding conducted by an independent qualified consultant(s) evaluating the economic, market, technical, financial, and management feasibility of the proposed project or operation in terms of its expectation for success as outlined in appendix A to subpart D of this part" (4).

Five components. All mandatory. A study that does four of them beautifully and skips the fifth is non-compliant.

Appendix A to Subpart D lists the factors under each component. There is an odd wrinkle here: USDA published Appendix A in the Federal Register. The text-identical companion appendix for REAP at 7 CFR Part 4280, Subpart B, Appendix D is machine-readable and is what most of us transcribe from (10).

We have published a full annotated transcription separately; the short version is:

Economic feasibility is a cost-benefit analysis. Five factors: the minimum inputs (labor, infrastructure, utilities, feedstocks, renewable resources) the project needs to operate; contracts in place and contracts still to be negotiated, including terms and renewals; environmental risks; the cost of the project relative to the increase in revenues or benefits; and the overall economic impact, including new markets and economic development.

Market feasibility covers current and future market potential, competition, and sales or service estimates including current and prospective buyers. Six factors: competition; the type of project (service, product, or commodity based); target market (new versus established); end-user analysis (captive versus competitive); by-product revenue streams; and industry risk.

Technical feasibility covers the reliability of the technology and the delivery of goods or services, including transportation, location, materials, and labor. Nine factors: commercial availability; the product or process success record and whether results can be duplicated; the experience of service providers; road, rail, and airport infrastructure; the need for local transportation; the labor market; availability of materials; the use, age, and reliability of the technology; and construction risk.

Financial feasibility covers the ability of the project to generate sufficient income, credit, and cash flow to sustain itself long term and meet all debt obligations. Twelve factors, and this is where USDA reads most closely: commercial or project underwriting; management's assumptions; accounting policies; source of repayment; dependency on other entities; equity contribution; market demand forecast; peer industry comparison; cost-accounting system; availability of short-term credit; adequacy of raw materials and supplies; and sensitivity analysis.

Management feasibility covers the legal structure, ownership, board, and management. Five factors: history of the business or organization; professional and educational background; experience; skills; and the qualifications needed to implement the project.

That is thirty-seven factors. The appendix also requires an executive summary at the front (nature and scope, purpose, location, design, capacity, estimated capital costs, and a summary of the feasibility determination for each component) and a recommendation plus a statement of the author's qualifications at the back (10).

Two of those factors deserve a note. Sensitivity analysis is an enumerated financial factor, which means a study without base, downside, and upside cases is facially deficient, not merely weak. And management feasibility is, in my experience, the component consultants treat as a formality and State Offices increasingly do not.

The study is not the business plan, and it is not the appraisal

Borrowers regularly assume that one of three other documents can do the study's job. None of them can.

The business plan. The B&I lender checklist says a business plan should be submitted "unless the information is contained in the feasibility study or in the credit evaluation," and may be omitted for a pure debt refinance (11). So the study can absorb the business plan. The reverse is not true. When a study is required, the borrower's own plan and projections cannot replace it, because the entire point of the requirement is that someone without a stake in the outcome tested the numbers.

The appraisal. The appraisal establishes collateral value. The study tests whether the projected cash flow will actually show up. For going-concern assets like hotels, senior living, car washes, and gas stations, the two have to be coordinated, because under 5001.203 any value the appraiser attributes to the business or the going concern has to be deducted from market value before the collateral discount is applied, and the appraiser has to physically visit an existing facility (12). The cash flow question belongs to the study; the value allocation belongs to the appraiser. And, again, the income approach of an appraisal is not a feasibility study.

The lender's credit evaluation. Under 5001.202, the lender has to produce its own written credit evaluation, and under 5001.202(b)(6)(ii) that includes a separate written evaluation of the feasibility study (13). Incorporating the study by reference is not enough and is a common reason files get sent back. The 2024 amendment also added 5001.202(b)(6)(iv), which says that "financial projections deviating from historical financial performance must be substantiated and documented" and that increases to revenues, margins, or profitability "should be reasonable and substantiated in the analysis" (13). That language is aimed at lenders, but the study is where the substantiation lives.

Why USDA cares this much

A regulation like this does not come from nowhere. It comes from losses.

In 1999 the General Accounting Office reviewed B&I guaranteed loans made between fiscal 1994 and 1998 on which USDA had paid claims. Of 24 borrowers examined, USDA had failed to follow its own requirements on 18. Eleven of those loans were missing feasibility studies that should have been obtained. Three more had studies with significant flaws (14). Field staff told GAO they had skipped the study when the lender's information "appeared to provide an adequate basis." GAO recommended that USDA clarify when studies must be obtained and train staff to tell a feasibility study from a business plan. USDA did both, via an administrative notice in January 2000 and nationwide training. The bright-line $1 million trigger you are reading about is, in a real sense, the descendant of that audit.

The modern echo is louder. The Office of the Comptroller of the Currency published a Community Developments Insights report on the B&I program in June 2025 that contains the only public series on application throughput. In fiscal 2021, 369 of 414 applications were approved, roughly 89 percent. In fiscal 2022 it was 367 of 590, roughly 62 percent. In fiscal 2023 it was 314 of 596, roughly 53 percent (15). Two caveats that most people who quote those numbers leave out: submissions jumped 44 percent over that period while USDA's capacity did not, so part of the decline is rationing rather than credit tightening; and the numbers are OCC's compilation of applications and approvals in the same fiscal year, which are not necessarily the same cohort. The direction is not in doubt.

Then came 2026. On February 18, 2026, Rural Business-Cooperative Service Administrator J.R. Claeys wrote an open letter to the more than 775 lenders in the OneRD program. The portfolio had passed $12 billion in active guarantees. It also carried, in his words, "over $1 billion in delinquent loans within the portfolio," and the agency had "paid approximately $300 million between repurchases and losses over the last year." The letter laid out ten underwriting expectations (realistic monthly cash flow projections for ramp-up and interest-only periods, working capital sized against industry norms, off-take contracts aligned with loan term, reserves funded from borrower equity rather than the loan, among others) and warned that negligent underwriting would lead to removal from the program under 7 CFR 5001.132 (16).

On May 12, 2026, USDA did exactly that, revoking the approved-lender status of ten lenders whose portfolios held roughly $620 million in delinquent loans, about 47 percent of Rural Development's total delinquencies (17). The revocations followed desk audits citing significant non-compliance with OneRD requirements.

I do not think it is a coincidence that the vintages now going bad are the fiscal 2021 through 2024 loans made during the program's fastest growth, when average loan size rose from under $4 million to over $8 million (18). Bigger loans, more of them above the study threshold, a lot of them to new businesses in a hurry.

The point for a borrower in 2026 is simple. The feasibility study is no longer a box to tick on the way to a guarantee. It is a loss-defense document, for the lender as much as for USDA, and it is being read that way.

Why studies get rejected

USDA reviews the study during the conditional commitment stage. The State Office reviews first; larger transactions above the State Director's delegated authority go to the National Office. Studies get sent back, or get the whole file sent back, for a fairly consistent set of reasons. The first three come straight from the codified expectations; the rest are practice.

1. Generic market analysis. Recycled industry overviews, an undefined trade area, demand estimated from total population rather than the actual demand cohort, and no accounting for the competitive pipeline. This fails the market component on its face.

2. Projections without an assumption sheet, or that ramp aggressively without support. The lender checklist requires that "financial projections must be supported by a list of assumptions showing the basis for the projections," and 5001.202(b)(6)(iv) now requires substantiation of any deviation from history (11, 13).

3. No sensitivity analysis. An enumerated Appendix A factor. Its absence is a facial deficiency.

4. Independence or qualification failures. A contingent fee, a packager relationship, or a CV that shows no experience in the relevant asset class.

5. Stale data. Under 5001.303, the borrower's financial statements must be dated within ninety days of the complete application (3). Market and demographic data more than a year old draw questions.

6. A thin management section. The most frequently underdeveloped component. Two paragraphs of biography is not a management feasibility analysis.

7. A perfunctory lender evaluation. The lender's separate written evaluation of the study under 5001.202(b)(6)(ii) is missing or consists of "we reviewed the study and concur."

   
   

There is also a category of studies that fail before they are read: the $3,000 template. A few firms sell what amounts to a fill-in-the-blanks document. State Offices have seen enough of them to recognize the format, and the National Office in particular has a low tolerance for them. Paying for one is not cheaper than paying for a real study; it is more expensive, because you end up paying twice.

What it costs and how long it takes

USDA does not publish a fee schedule, so every number here is market-reported and comes from firms that sell studies, including us. Treat them accordingly.

For a single-purpose B&I project, most quotes fall between $10,000 and $25,000. Studies for complex projects with multiple unrelated revenue lines, biorefineries, or large food processing plants run higher, with $30,000 to $50,000 quoted by several firms. Some engagements start lower; our fixed-fee engagements start at $4,900 for the simplest scopes. Timelines range from ten business days on the fast end to four to eight weeks on the slow end (19, 20).

Two practical notes. Cost scales with the number of distinct major revenue lines, because a winery with an events venue is, for the market and financial components, effectively two studies. And the study cost can generally be financed into the B&I loan as a professional-service or pre-development cost, which takes some of the sting out for the borrower.

What changed in 2024, 2025, and 2026

The rule you are reading is not the rule that existed in 2020, and a lot of the pages competing for your attention were written against the old text.

September 30, 2024 (89 FR 79698), effective November 29, 2024. The most substantive amendment since OneRD launched. It replaced the original "new entity" and "existing entity" language in 5001.306 with "new business" and "existing business" and added the matching definitions to 5001.3. It rewrote parts of the lender's credit evaluation, including the substantiation requirement for projections at 5001.202(b)(6)(iv). It added the going-concern deduction and the physical site visit requirement for appraisals at 5001.203. The dollar thresholds did not change (21). A correction followed on December 9, 2024 (89 FR 97477), and USDA confirmed the November 29, 2024 effective date in a December 12, 2025 notice.

August 19, 2025. Implementing Executive Order 14315, USDA announced that "for the USDA Rural Development Business and Industry (B&I) Guaranteed Loan Program, wind and solar projects are not eligible," effective immediately (22). This was done by policy announcement and Administrator guidance rather than by amending Part 5001, so the CFR text does not reflect it. Any B&I project whose revenue model leaned on wind or solar generation needs to be re-scoped before anyone spends money on a study.

December 11, 2025 (90 FR 57351). Technical corrections revising the "affiliate" definition to align with SBA's rule at 13 CFR 121.301(f) and restoring the "commercially available" definition that had been dropped by mistake.

March 9, 2026 (91 FR 11272). The FY2026 annual fee and guarantee notice, retroactive to October 1, 2025. For the first time, B&I guarantee percentages are tiered by size: 85 percent for loans under $5 million, 80 percent for $5 million and above, with the 3 percent upfront fee and 0.55 percent annual retention fee unchanged (5). This does not change the feasibility rules, but it improves lender economics in exactly the $1 million to $5 million band where the mandatory study lives.

February and May 2026. The Claeys letter and the ten lender revocations, covered above.

What to actually do

If you are a borrower or a lender with a B&I deal on the desk, here is the sequence I would follow.

Classify the borrower first. Under a year old, not yet stabilized, or a new affiliate entering a new market: new business. Over a year, stabilized, same entity, no significant change in operations: existing. If it is close, ask the State Office in writing.

Measure the guaranteed loan, not the project. Over $1 million and new business: commission the study now, not after the lender's package is drafted. At or under $1 million, or existing business: build a lender analysis strong enough that USDA does not feel the need to ask, and be ready to commission one if it does.

Pick a consultant for independence and asset-class fit. No packaging, brokerage, ownership, or contingent-fee relationship. Documented experience in the specific property type. Confirm acceptability with the State Office before engagement if the consultant is unknown to them.

Build to Appendix A explicitly. All five components, every enumerated factor addressed, an assumption sheet, three sensitivity cases, an executive summary with a determination for each component, a recommendation, and the author's qualifications.

Coordinate the study with the appraisal on going-concern assets so the two documents do not tell different cash flow stories.

Refresh the financials so the balance sheet and year-to-date income statement are within ninety days of the complete application.

Have the lender write its own evaluation of the study. Not a paragraph. An evaluation.

Read the Claeys letter and check the deal against its ten points, especially any interest-only or ramp-up period, working capital sizing, and reliance on speculative off-take or by-product revenue. Those are the items USDA has said it is looking for.

September 16, 2026 by Michal Mohelsky, principal of MMCG Invest, LLC, a national SBA and USDA feaisbility study consultancy

Request a Feasibility Study https://calendar.app.google/EJzWEz3GCqLY2jU86

Michal Mohelsky, J.D. | Principal | mmcginvest.com 

Contact: michal@mmcginvest.com

Phone: (628) 225-1125

Frequently asked questions

Is a feasibility study always required for a USDA B&I loan? No. It is mandatory only for guaranteed loans over $1,000,000 to a new business (5001.306(a)(3)(i)). Below that, or for existing businesses, USDA may require one at its discretion.

Is the $1 million measured on the project or the loan? On the guaranteed loan amount. Not total project cost, and not the guaranteed portion.

What is the $600,000 threshold I have seen mentioned? It is the line between the streamlined application package under 5001.306(b) and the full package under 5001.306(a). It has nothing directly to do with feasibility studies, which is why most consultant pages leave it out, but it determines which set of application requirements you are working under.

My borrower has been operating for fourteen months. Are they automatically an existing business? No. More than one year is necessary but not sufficient. The business also has to have reached full operational capacity or stable operations as determined by the Administrator.

Can the lender or the borrower's accountant write the study? No. It has to be prepared by an independent qualified consultant. Anyone with a financial interest in the loan, the borrower, or the project fails the conflict of interest test.

Is there a list of USDA-approved consultants? No. Acceptability is decided by the State Office reviewing the file, based on credentials, relevant experience, and absence of conflicts.

Can the study be waived on a strong deal above $1 million? No. The requirement is mandatory; only the scope is adjustable. The only route out is establishing that the borrower is an existing business.

Can the business plan substitute for the study? No. The study can absorb the business plan, but not the other way around.

Can the study cost be financed into the loan? Generally yes, as a professional-service or pre-development cost. Confirm with your lender.

Does the same rule apply to REAP or Community Facilities? Each program has its own application section in Part 5001 (5001.304 for Community Facilities, 5001.307 for REAP), with different triggers and, for REAP, a technical report requirement. The $1 million new business trigger described here is the B&I rule at 5001.306.

Are wind and solar projects still eligible for B&I? Not as of August 19, 2025, by USDA policy announcement. Check the current status before scoping.

What is 7 CFR 5001.306? The section of the OneRD guaranteed loan regulation that sets out the B&I-specific application requirements, including the feasibility study rule.

The reason so many pages get this wrong is that the rule is short and the definitions are long, and it is easier to write "over $1 million" than to explain what "guaranteed loan" and "new business" mean. But the definitions are where the money is. A borrower who understands that a new acquisition entity is a new business, that the threshold is the loan and not the project, and that there is no waiver above the line, will structure the deal and budget for the study before the lender asks. That borrower closes. The one who found out in the conditional commitment stage usually does not, or does so six months late.

If you want a second opinion on whether your deal trips the trigger, we will give you one for free, and we will tell you if the answer is that you do not need us.

Sources

1. 7 CFR 5001.306, Specific application requirements for B&I projects. eCFR, current through September 2026. https://www.ecfr.gov/current/title-7/subtitle-B/chapter-L/part-5001/subpart-D/section-5001.306

2. OneRD Guaranteed Loan Regulation, final rule, 85 FR 42494 (July 14, 2020), effective October 1, 2020; Appendix A images at 85 FR 42518. https://www.govinfo.gov/content/pkg/FR-2020-07-14/html/2020-13991.htm

3. 7 CFR 5001.303, General application requirements, including 5001.303(c)(4). eCFR.

4. 7 CFR 5001.3, Definitions (feasibility study, new business, existing business, qualified consultant, conflict of interest, guaranteed loan). eCFR. https://www.ecfr.gov/current/title-7/subtitle-B/chapter-L/part-5001/subpart-A/section-5001.3

5. OneRD Annual Notice of Guarantee Fee Rates, Periodic Retention Fee Rates, Loan Guarantee Percentage for Fiscal Year 2026, 91 FR 11272 (March 9, 2026). https://www.federalregister.gov/documents/2026/03/09/2026-04581/

6. 7 CFR 5001.406, Guaranteed loan amounts. eCFR.

7. USDA Rural Development, OneRD Guarantee Loan Initiative Lender Guide. https://www.rd.usda.gov/media/file/download/onerdlenderguide.pdf

8. 7 CFR 5001.208, Conflicts of interest. eCFR.

9. 7 CFR 4279.150 (legacy B&I feasibility provision): "The income approach of an appraisal is not an acceptable feasibility study."

10. 7 CFR Part 4280, Subpart B, Appendix D, Feasibility Study Components (text-identical to Part 5001 Appendix A). Cornell LII. https://www.law.cornell.edu/cfr/text/7/appendix-D_to_subpart_B_of_part_4280

11. USDA Rural Development, B&I Lender's Complete Application Checklist. https://www.rd.usda.gov/media/file/download/bi-lendercompleteapplicationchecklist.pdf

12. 7 CFR 5001.203, Appraisals. eCFR.

13. 7 CFR 5001.202, Lender's credit evaluation, including (b)(6)(ii) and (b)(6)(iv). Cornell LII. https://www.law.cornell.edu/cfr/text/7/5001.202

14. GAO/RCED-99-249, Rural Development: Rural Business-Cooperative Service Business Loan Losses (August 1999).

15. Office of the Comptroller of the Currency, Community Developments Insights, USDA Rural Development Business and Industry Guaranteed Loan Program (June 2025). Application and approval counts FY2021 to FY2023.

16. USDA Rural Business-Cooperative Service, open letter from Administrator J.R. Claeys to OneRD lenders, February 18, 2026 (posted as USDA Stakeholder Announcement February 20, 2026).

17. USDA press release, USDA Revokes Approved Lender Status of Ten Lenders, May 12, 2026.

18. Summit LLC for the National Rural Lenders Association, USDA B&I Guaranteed Loan Program: Economic Assessment 2025, entered into the U.S. House Agriculture Committee record September 18, 2025 (loan counts and average loan size, FY2015 to FY2024).

19. USDA-feaisbility-study.com USDA Feasibility Study Costs 2026 (market-reported pricing).

20. UnionMetric Appraisal & Feasibility, USDA Business and Industry Feasibility Study Guide 2026 and cost guide (market-reported pricing and timelines).

21. OneRD Guarantee Loan, final rule, 89 FR 79698 (September 30, 2024), effective November 29, 2024; corrections at 89 FR 97477 (December 9, 2024); effectiveness confirmed December 12, 2025. https://www.federalregister.gov/documents/2024/09/30/2024-21920/onerd-guarantee-loan

22. USDA press release, Secretary Rollins Blocks Taxpayer Dollars for Solar Panels on Prime Farmland, August 19, 2025; Executive Order 14315, 90 FR 30821 (July 10, 2025).

23. OneRD Guaranteed Loan Regulation, technical corrections, 90 FR 57351 (December 11, 2025).