Conventional Bank Loan Method: Appraisal Rules, Supervisory LTV Limits and the Study a Bank Reads

Where the Rules for a Conventional Commercial Loan Come From

A conventional commercial real estate loan carries no federal guarantee, so no agency program stands between the borrower and the bank. The SBA and USDA rules tell a lender what the government will accept before it shares the risk. A bank lending on its own balance sheet answers instead to its own written lending policy, and that policy is shaped by three sets of federal rules: the real estate lending standards with their supervisory loan-to-value limits, the appraisal regulation, and the capital rule's treatment of high volatility commercial real estate. This article walks through each one, explains what it asks of a bank, and marks where a feasibility study supplies evidence the credit memo needs and where a study has no role at all.

The starting point is a short regulation. Every FDIC-supervised bank must adopt and maintain written policies that set limits and standards for credit secured by real estate ^[1]. National banks carry the same obligation under the OCC's version of the rule ^[2], and the Federal Reserve's version sits in 12 CFR part 208; the agencies adopted the text as one uniform rule published in three places ^[3]. The board must review and approve the policy at least once a year, and the policy must contain diversification standards and underwriting standards, loan-to-value limits among them, that are clear and measurable ^[4]. The bank must also keep watching the real estate market where it lends and keep the policy in step with it ^[5].

Two consequences follow for anyone preparing or reading a study for a conventional loan. First, the numbers that decide the loan (the minimum coverage, the maximum loan to cost, the internal loan-to-value limit, the required equity) are the bank's own numbers, set in its own policy rather than in any public rule. No federal regulation sets a conventional coverage floor. Second, the federal rules draw the outer boundaries and name the questions the policy has to answer, so a study organised around those questions follows the structure the policy itself is expected to have ^[6]. The archived post Feasibility Studies, Business Plans, and Investment Memos: What Banks Really Need compares the documents a bank may ask a borrower for; this article takes the regulatory route instead, one rule at a time. A reader whose loan carries an SBA guarantee will find the regulation read test by test in this series' guaranteed lending counterpart, built on 13 CFR 120.

The Questions the Interagency Guidelines Put to a Lending Policy

The rule is brief; the substance sits in its Appendix A, the interagency guidelines that every bank's real estate policy is expected to reflect. The underwriting discussion opens with the credit factors a prudently underwritten loan should reflect: the capacity of the borrower, or the income of the property, to service the debt, the value of the mortgaged property, the borrower's overall creditworthiness and the equity invested ^[7]. The list goes on to secondary repayment sources and credit enhancements; this article concentrates on the first four, because they are where a study's evidence bears. Capacity and equity, and the market evidence behind both, are questions a feasibility study is built to answer. Value belongs to the appraiser, and that division runs through everything that follows.

Construction and development lending, and loans on finished commercial buildings, get a longer list, scaled to the size and type of each project ^[8]. Its first item matters most here: the policy should set out the conditions that call for a feasibility study and the sensitivity and risk analysis that accompanies one, and the example the appendix gives is how projected income responds to interest rates, vacancy rates or operating expenses ^[9]. The same list asks for minimum hard equity, meaning cash or an unencumbered investment in the property, and for floors on the net worth and cash flow of the borrower or the property and on the coverage of debt service ^[10]. It also asks for standards on interest reserves and for pre-leasing or pre-sale requirements on income-producing property ^[11]. The 2006 interagency guidance on commercial real estate concentrations repeats the point in its own underwriting list, placing feasibility studies and stress testing beside hard equity ^[12].

The guidelines then name the market evidence a bank should weigh. The supply and demand factors include population and employment trends, zoning, and current and projected vacancy, construction and absorption ^[13]; current and projected lease terms, rents and sale prices, concessions included ^[14]; and operating expenses by type of project, the economic diversification of the lending area, and valuation trends such as discount and capitalization rates ^[15]. Read together, the two lists describe the work of a market and financial feasibility study nearly line for line: who will rent or buy, at what price, against what competing supply, at what cost to operate, and what happens to coverage when any one of those numbers moves.

Nothing in the appendix fixes a coverage ratio, an equity percentage or a range of stress. The bank writes those numbers, and a study should test against them rather than against a figure borrowed from a guaranteed program. Where coverage is discussed below, the conventions of the series' worked examples apply.

Conventions. This article follows the conventions of the Loan Analytics worked examples. Coverage is tested against the program floor the site's program page states, where the program has one, and against a stated 1.25x lender assumption, and the higher of the two binds. Equity is measured at the loan the cash flow supports, with fees recomputed on that loan. An acquisition is tested on the seller's historical earnings, not on projections.

A conventional loan has no program floor to test against, so the stated lender assumption is the only threshold in play in this article's arithmetic. A study for a real loan should replace that assumption with the bank's actual policy figure as soon as the bank discloses it, and say which figure it used.

The Supervisory Loan-to-Value Limits, Category by Category

The loan-to-value limits are easy to misread. They are not the levels a bank lends to. They are ceilings on the internal limits each bank sets for itself, and the internal limits should not exceed them ^[16]. Table 1 sets out the categories as the appendix prints them.

Which row applies turns on the definitions at the end of the appendix. A construction loan finances the erection or rehabilitation of a building or other structure, with the infrastructure it needs ^[21]. The improved property definition lists five kinds of collateral. Three are residential property of five or more units, completed commercial property, and other income-producing property that is complete and available for occupancy and use ^[22]. The other two are farmland, ranchland or timberland in ongoing agricultural production, and 1- to 4-family residential property that its owner does not occupy ^[23], and income-producing 1- to 4-family property that the owner occupies is excepted ^[24]. So a new apartment building sits under the 80 percent construction row while it is being built ^[18], and the finished building of five or more units falls under the 85 percent improved property row ^[19]. Of the three conventional worked examples linked at the end of this article, the two new-construction projects, the garden apartment complex and the short-term rental cabins, begin under a construction row before they reach a permanent one; the marina example is the purchase of an operating property and is tested at the improved property limit.

When one loan funds several phases, such as land development followed by vertical construction, the limit of the final phase governs, but the bank should not disburse more than the development or construction outlays actually made ^[25]. That second clause is the reason a construction budget and a draw schedule belong in any study that supports a construction loan. The ceiling is computed on value; the money moves on cost.

The same passage adds two cautions. A lender should consider a lower internal limit for any subcategory that carries more risk than its category as a whole ^[26]; a single-tenant special purpose building is the kind of subcategory a policy could single out under that sentence, against a multi-tenant warehouse in the same category. And the guidelines say in terms that a loan at the supervisory levels is not, for that reason, a sound loan ^[27]. A loan at 80 percent of value with thin coverage is a thin loan ^[18]. The ratio is one credit factor among several, and the appendix treats it that way.

The ratio is also only as good as the value beneath it. In the guidelines, value means an opinion or estimate of market value set out in an appraisal or an evaluation prepared under the appraisal regulations ^[28]. When the loan buys an existing property, the guidelines set value at "the lesser of the actual acquisition cost or the estimate of value" ^[29]. The rule matters most on a purchase where the appraisal comes in above the contract price, and Table 6, in the sizing section below, shows the arithmetic: the limit is applied to the price, not to the higher appraised figure, and a capital stack built on the appraisal alone overstates the loan the bank can make.

Loans Above the Limits, Loans the Limits Exclude, and the Concentration Screens

A bank may lend above the supervisory limits in individual cases where other credit factors support the loan ^[30], and it should record such loans and report their total to the board at least quarterly ^[31]. The guidelines also put the exceptions inside a capital budget. In aggregate, loans above the supervisory limits should not exceed 100 percent of the bank's total capital ^[32]. Within that total, the share secured by commercial, agricultural, multifamily or other property outside the 1- to 4-family class should not exceed 30 percent of total capital ^[33]. Each exception should be approved on a written justification that sets out every relevant credit factor behind the decision ^[34].

That justification is where an independent study can earn its place in an exception file. The credit factors it can draw on are the ones the guidelines list for every loan, among them the capacity of the borrower or the income of the property to service the debt, the borrower's creditworthiness and the equity invested ^[7], and for income property the pre-leasing or pre-sale position ^[11]. A study written to support an exception should document absorption, achievable rents and stressed coverage, so that the justification can name those factors in a form an examiner can read later. It cannot make an exception prudent, and it should not be written as though it could.

Some loans sit outside the limits altogether. The guidelines list a set of excluded transactions, and the first is directly relevant to the comparison with guaranteed lending: a loan guaranteed or insured by the U.S. government or one of its agencies is excluded when the guaranty is at least as large as the portion of the loan above the limit ^[35]. A guaranteed loan and a conventional loan on the same building are therefore measured against different ceilings, and loan-to-value figures drawn from guaranteed lending say little about what a conventional bank may do.

Above the individual loan sits the portfolio. The 2006 interagency guidance on concentrations names two screens that prompt closer supervisory review: construction, land development and other land loans at 100 percent or more of total capital ^[36], or total commercial real estate loans at 300 percent or more of total capital where that portfolio has grown by 50 percent or more over the prior 36 months ^[37]. The guidance is explicit that the screens are not caps on lending ^[38]. Because the screens concern the bank and not the project, a borrower should ask early whether the bank has room for the loan before paying for the work that supports it. A study does not change a bank's concentration position; it can only make the individual credit easier to explain.

When an Appraisal Is Required, and When an Evaluation Will Do

The appraisal regulation for FDIC-supervised banks is 12 CFR part 323; the OCC's sits in 12 CFR part 34 and the Federal Reserve's in 12 CFR part 208 and part 225 ^[39]. The general rule is wide: an appraisal by a state certified or licensed appraiser is required for every real estate-related financial transaction unless an exemption applies ^[40]. A commercial real estate transaction is any such transaction not secured by a single 1- to 4-family residential property ^[41]. Table 2 lists the exemptions and thresholds that matter most for commercial lending.

Rows 2 and 3 reach only federally related transactions, and the regulation defines those as transactions that require the services of an appraiser ^[49]. A loan covered by an exemption is therefore outside them: a business loan within row 4 can exceed the $500,000 commercial figure ^[44] and still need no appraisal while it stays within the $1 million limit ^[46], and a guaranteed transaction in row 6 needs none under the part at any size ^[48]. The OCC's rule carries the same commercial threshold of $500,000 for national banks ^[50]. Three further points in the table are easy to miss. The business loan exemption depends on the source of repayment, and the interagency appraisal guidelines expect the bank to establish that the primary source is the operating cash flow of the business, not rent or a sale of the real estate ^[51]. An investor property leased to third parties, whose repayment is the rent, never fits it ^[46]. The exemption for guaranteed transactions removes the part 323 requirement, but not whatever valuation the guaranteeing agency demands; SBA, for one, can make professional appraisals or a feasibility study a condition of its guaranty ^[52]. And an exemption from an appraisal is not an exemption from valuation: for the residential, business loan, existing credit, small commercial and rural residential exemptions, the bank must still obtain an appropriate evaluation of the collateral ^[43], and the FDIC may require a full appraisal whenever it judges one necessary for safety and soundness ^[53].

The regulation also sets minimum standards for every appraisal in a federally related transaction. It must conform to the Uniform Standards of Professional Appraisal Practice ^[54]. It must be written and must carry enough information and analysis to support the bank's decision to make the loan ^[55]. It must analyse and report deductions and discounts for proposed construction or renovation, partially leased buildings, leases at non-market terms and tract developments with unsold units ^[56]. And a fee appraiser must be engaged directly by the bank or its agent and have no interest in the property or the transaction ^[57]. The interagency guidelines extend the same independence to evaluations: the bank should select and engage the evaluator itself, and independence is compromised when a borrower recommends the appraiser ^[58]. The archived guide How to Review an Appraisal Report: A Step-by-Step Guide covers the reviewer's side of that file.

Why a Feasibility Study Is Not an Appraisal

The regulation defines the appraisal precisely, and the definition is the clearest way to see what a feasibility study is not. Under 12 CFR 323.2(a), an appraisal is a written opinion of market value that a qualified appraiser prepares independently and impartially, for an adequately described property as of a specific date, supported by the presentation and analysis of relevant market information ^[59]. Market value has its own definition: the most probable price the property should bring in a competitive and open market under the conditions of a fair sale, with buyer and seller each acting prudently and knowledgeably ^[60]. A feasibility study answers a different question. Its question is whether the project's market, its costs and the income it will earn can support the loan under review, and it takes the appraised value as an input handed to it, never as a conclusion of its own.

The distinction is not a matter of modesty. For a construction or renovation loan, the bank would generally ask the appraiser for the market value as is and, where they apply, prospective market values upon completion and upon stabilization ^[61]. The as completed value is the market value at the time development is expected to finish, and the as stabilized value is the market value when the property is projected to reach stabilized occupancy ^[62], which the guidelines describe as the level a property should reach after reasonable exposure to the market on terms like those of similar properties ^[63]. The bank should understand the as is value and weigh the prospective value that matches the credit decision and the phase being funded ^[64], and the HVCRE capital test in the next section runs on the as completed figure as well. Wherever an appraisal is required, that value comes from a state certified or licensed appraiser ^[40]. A fee appraiser must be engaged directly by the bank or its agent, never by the borrower ^[57], and the bank may also accept an appraisal prepared by an appraiser that another financial services institution engaged directly, if the rule's conditions are met ^[65]. A study that printed a value conclusion of its own would only invite confusion with that figure.

The interagency appraisal guidelines also say what market work can and cannot do inside the valuation file. The passage is written about housing: "information on local housing conditions and trends, such as a competitive market analysis, does not contain sufficient information on a specific property that is needed, and therefore, would not be acceptable as an evaluation" ^[66]. Its reason, that market information says too little about the specific property, applies just as well in this article's reading to a commercial market study, though the guidelines do not say so in terms. The same passage adds that such information, though insufficient as an evaluation, may help in developing one or in developing an appraisal ^[67]. On that reading, a study's rent survey, competitive supply count and absorption schedule are evidence an appraiser may read and test; they are never a substitute for the appraiser's own opinion of value. The guidelines make a parallel point about proposed developments: an appraiser may be asked for the sum of the retail sales of the units, but that sum is not the property's market value under the regulations ^[68].

There is one place where the guidelines name the feasibility study outright inside the valuation rules. Where a bank finances homes or a small condominium building one unit at a time, it may rely on individual unit appraisals if it can show, through an independently obtained feasibility study or market analysis, that every unit securing the loan can be built and sold within 12 months ^[69]. The study there supports the bank's choice of valuation method; it still does not value the units.

The appraisal side also explains why a study's absorption and lease-up analysis must stand on its own evidence. For proposed and partially leased rental developments, the appraiser must deduct and discount for the fact that the property has not reached stabilized occupancy, and the analysis should also consider how the unleased space will be absorbed ^[70]. If the study and the appraisal reach the same lease-up period, a reader of both should ask whether they reached it independently or whether one copied the other. A study should therefore state its own sources for absorption and rent, date them, and note where its conclusion differs from the appraisal and why. The byline of every article in this series ends with the words Not an appraisal for the same reason.

High Volatility Commercial Real Estate and the Capital a Bank Holds

The capital rule weighs some construction loans more heavily than others. Under 12 CFR 324.2, the definition of a high volatility commercial real estate (HVCRE) exposure starts from "A credit facility secured by land or improved real property", for as long as the bank has not reclassified it ^[71]. Such a facility counts when it "Primarily finances, has financed, or refinances the acquisition, development, or construction of real property" ^[72]. Its purpose must be to turn that property into income-producing real property ^[72], and its repayment must rest on the property's future income, its sale proceeds or a refinancing of it ^[73]. The definition then carves out several kinds of financing, among them community development property and agricultural land ^[74]. An FDIC-supervised bank must assign such an exposure a risk weight of 150 percent ^[75], against the 100 percent weight that applies to corporate exposures generally ^[76]. For the same exposure, the higher weight produces half again the risk-weighted amount the bank must hold capital against. How a given bank reflects that in its terms is the bank's decision, and a study should not assume an answer; what a study can do is set out the facts that decide whether the definition applies.

The definition carries its own exits. A commercial real property project escapes it only when every one of three conditions holds. The first is that the loan-to-value ratio stays within the supervisory maximum for its category ^[77]. The second condition reads: "The borrower has contributed capital of at least 15 percent of the real property's appraised, 'as completed' value to the project" ^[77]. That capital counts only as cash, unencumbered readily marketable assets, development expenses paid out of pocket, or contributed real property or improvements ^[78]. The third condition is timing: the capital must be in the project before the bank advances funds ^[79] and be contractually required to stay there until the exposure is reclassified ^[80]. Buying or refinancing an existing income-producing property, with a mortgage on it, also falls outside the definition, provided the income the property already earns is enough to pay both what it owes the lender and what it costs to run, measured against the bank's own standards for a permanent loan ^[81]. And a bank may reclassify a construction exposure once the project is substantially complete and its cash flow supports its debt service and expenses ^[82].

Each exit is a question of evidence a study can help assemble. The capital test needs a sources and uses table that shows what the sponsor has put in, in what form and when, measured against the appraiser's as completed value. The reclassification test needs a defensible date for substantial completion and a lease-up or ramp schedule that shows when stabilized cash flow will cover the debt. Neither test is decided by the study: the bank applies its own underwriting criteria for permanent loans, and the appraiser supplies the value. But a sponsor who knows the 15 percent figure before the capital stack is fixed ^[77] can avoid a structure that pushes an otherwise ordinary loan into the higher capital class.

Sizing the Loan: Three Bank Tests, One Supervisory Ceiling and the Stresses

The rules above meet in one piece of arithmetic: the loan amount. In a study's model of a construction-to-permanent loan on an income property, the bank's policy limits are run side by side and the lowest result is taken as the loan, since the loan must clear every one of them. The illustration below uses made-up project figures, every one labelled as an assumption, with four sourced inputs: the 10-year Treasury yield from the Federal Reserve's H.15 release ^[83], the supervisory limits for commercial construction ^[18] and for improved property ^[19], and the 15 percent capital figure from the HVCRE definition ^[77]. The appraised value is an input the bank's appraiser would supply, entered here as an assumption; the illustration does not estimate it. Table 3 lists every input.

Rounded as shown; lines may not foot by one unit in the last place.

On these assumptions, set out in Table 4, the project produces net operating income of $788,500 at stabilization, and the permanent rate of 7.11% gives an annual debt service constant of 8.57%. The bank's loan-to-cost limit allows $7,500,000 and its internal loan-to-value limit allows $8,400,000, but the coverage test, income divided by the assumed 1.25x lender minimum and by the constant, allows only $7,364,219, and that is the loan the cash flow supports. The supervisory ceiling of $8,960,000 at the construction limit never comes into play; it is printed only to show how far below it a coverage-sized loan can fall. At the supported loan, loan to as completed value is 65.8%, loan to cost is 73.6%, the debt yield is 10.7% and coverage sits exactly at the assumed 1.25x minimum. The borrower's equity is $2,635,781, which is 23.5% of the as completed value and comfortably above the $1,680,000 the HVCRE exclusion would require, provided it goes in before the first advance.

On these assumptions coverage binds, and only narrowly: at any lender minimum below 1.23x the loan-to-cost limit would bind instead, so the result depends on the assumed 1.25x lender minimum. Here the resulting loan-to-value ratio looks low only because income, not collateral, sets the loan. A study should therefore report the coverage test and the binding limit beside the loan-to-value ratio, never the ratio alone.

The guidelines name three variables for sensitivity analysis: interest rates, vacancy and operating expenses ^[9]. Table 5 holds the loan at the supported amount and moves each variable in turn, then all three together.

A rate two points higher at conversion or reset, 9.11% in place of 7.11%, lifts debt service to $748,271 and cuts coverage to 1.05x. Vacancy at 12.0% in place of 7.0% cuts net operating income to $716,000 and coverage to 1.14x. Operating expenses up 10.0% cut income to $732,500 and coverage to 1.16x. With all three together, income of $660,000 against debt service of $748,271 gives 0.88x, and the property would need 94.1% occupancy to cover its expenses and debt service, against 82.1% in the base case. Each single stress still leaves income above debt service, yet each one alone pushes the ratio under the assumed 1.25x lender minimum, so a loan sized to that minimum on the first day has no room for any of them. That is the information the sensitivity requirement exists to produce, and it is more useful to a credit committee than a base case presented alone.

The second illustration returns to the valuation rule for purchases. Table 6 applies the 85 percent improved property limit to a building bought for less than its appraised value ^[19].

Because a purchase is measured on the lower of price and appraisal ^[29], the supervisory ceiling is $5,100,000 on the price of $6,000,000, not $5,440,000 on the appraised $6,400,000, and a capital plan written on the appraisal alone would overstate the available loan by $340,000. A bank that applies the same definition of value to its internal limit meets the same gap, whichever limit binds.

What the October 2026 Loan Officer Survey Says About Standards Now

The Federal Reserve's Senior Loan Officer Opinion Survey is the public record of how banks say their credit standards changed over the preceding three months. The October 2026 survey drew responses from 61 domestic banks and 18 U.S. branches and agencies of foreign banks ^[84]. Its summary reports that banks tightened standards, on net, for every commercial real estate category it asks about: construction and land development, nonfarm nonresidential and multifamily ^[85]. Banks also reported weaker demand across all three categories ^[86]. Table 7 gives the distribution of answers from domestic banks on standards.

Most banks held their standards where they were; the tightening is a net figure, driven by the minority that moved. Construction lending shows the sharpest tilt, and demand there was also soft: 31.0 percent of domestic respondents saw moderately weaker demand for construction and land development loans and 6.9 percent moderately stronger ^[90]. For nonfarm nonresidential loans the weaker side was 24.6 percent against 14.0 percent stronger ^[91], and for multifamily 31.6 percent against 8.8 percent ^[92].

For a study, the survey changes emphasis rather than method. A study prepared while banks report tightening should not count on an exception to policy, on a thin coverage margin at conversion, or on a bank's comfort with a submarket facing heavy new supply. The appraisal guidelines treat the same market variables, including vacancy, absorption, rents, concessions and construction overruns and delays, as the changes that should prompt a bank to look again at its collateral ^[93]. A study prepared in a tightening market should spend more of its length on exactly those variables, run its stresses on them, and say plainly which of its conclusions depend on the current rate path.

The SBA Record for the Same Asset Classes, Read as a Benchmark Only

The Loan Analytics loan dataset holds no conventional loan outcomes. It is built from SBA's 7(a) and 504 loan files. A public loan-level source for conventional commercial real estate outcomes was probed and set aside for a later build, and bank call reports carry no individual loan and no property location, so neither feeds the dataset. What the dataset can offer a conventional lender is the SBA record for the same kinds of property, which is a different population lending on different terms. A guaranteed loan sits outside the supervisory loan-to-value limits when the guaranty covers the slice above the limit ^[35], and outside the part 323 appraisal requirement altogether ^[48], so the loans below were made under a different rulebook from the one this article describes. Table 8 prints the record for the asset classes of the three conventional worked examples: marina, multifamily, and hotel and motel as the nearest class to a short-term rental business.

Loan Data Behind This Article

Every figure in this table comes from the Loan Analytics loan dataset, analytics-loan-data freeze-3 (commit 360c856), and carries its key. SBA rates are approval-cohort floors and are withheld when fewer than 30 loans stand behind them.

How to Read the SBA Benchmarks Beside a Conventional Loan

Every rate in the block is an approval-cohort floor. It sets the loans charged off so far against the loans disbursed, for approvals made in FY2017 through FY2021, with loans still open left in the count, so any later loss can only push it up. The national 7(a) floor is 6.33 percent ^[97] and the 504 floor is 0.35 percent ^[98]; the approval-cohort floor article explains why a loss rate published on this basis can only understate. A rate also needs enough loans behind it to mean anything. The 504 marina cell counts 80 ^[100] approvals, yet the block shows no loss figure for it, because its measured cohort is under 30 loans and the dataset prints no rate on so few.

The classes also differ in how well SBA reaches them. The marina class counts 220 ^[94] approvals under 7(a) over ten fiscal years and a charge-off floor of 0.00 percent ^[94], a small operating-business record in which no loan in the measured cohort has yet been charged off. The hotel and motel class is large, with 8,942 ^[96] approvals under 7(a) and a charge-off floor of 0.99 percent ^[96]. Multifamily is the instructive case. An apartment owner leases every unit to others and operates from none of them, and SBA's eligibility rule treats a landlord of that kind as a passive business it will not finance, outside a narrow exception for Eligible Passive Companies ^[101]. The small count is consistent with that rule: the class holds only 204 ^[95] approvals under 7(a) over ten years, and its charge-off floor of 4.27 percent ^[95] describes a small and unusual set of borrowers rather than the apartment market a conventional bank lends into.

Used properly, the block answers a narrow question: how has this kind of operating business performed when financed under a federal guarantee? That is context for a study's discussion of business risk on a marina or a lodging property. It is not a conventional default rate, it is not an input to a bank's loss reserve, and a study should never present it as either. The limits of the SBA loan file are set out in their own article.

What a Bank's Credit Memo Takes From a Study

A bank's real estate lending policy should reflect consideration of the interagency guidelines ^[6], so a study that follows the guidelines' order can be read against that policy item by item. Table 9 maps the guideline items a study can support to the evidence it should carry, and marks the items it cannot decide.

Two habits make a study more useful in that file. The first is to separate what is sourced from what is assumed, line by line, so a credit officer can replace an assumption with the bank's own figure without rebuilding the model. The three conventional worked examples listed under the related pages below, a marina on a federal lake lease, a garden apartment complex and a group of rental cabins, work this way: each prints an inputs table in which the bank's coverage minimum and loan-to-cost figure are labelled as assumptions that sit inside the supervisory ceilings. The second habit is to state the value premise the study relied on and where it came from. If the appraisal is not yet in, the study should say that its loan sizing is provisional and show how the result moves if the as completed value comes in lower.

A study should also be clear about who commissioned it. Under the appraisal rules a fee appraiser is engaged directly by the bank or its agent, never by the borrower ^[57]; no comparable rule governs a feasibility study, which a sponsor may commission. That makes independence a matter of disclosure and method rather than regulation: the study should state its sources, its assumptions and its conflicts, and it should be willing to report that a project does not carry the requested loan.

Where Conventional Loan Files Go Wrong, and What to Check

The checks below look for mismatches between documents that each look sound alone. A lender reviewing a study, and a borrower commissioning one, can run them.

The value premise does not match the test. The guidelines tie the value a bank weighs to the credit decision and the phase being funded ^[64], and a purchase runs on the lesser of price and appraisal ^[29]. A study that sizes a permanent loan on the as completed figure when the decision turns on stabilized operations, or an acquisition on the appraisal alone, has used the wrong number.

The phase is misread. A loan that funds land development and construction takes the final phase's limit, but its draws are capped by outlays ^[25]. A study whose sources and uses show the loan funding more than the costs incurred at any draw has a structure the guidelines warn against.

Coverage is reported at one point only. The guidelines ask for sensitivity of income projections to rates, vacancy and expenses ^[9]. A single base case at the bank's minimum hides how quickly the loan fails, as the stress table above shows.

Equity arrives late or in the wrong form. The HVCRE exclusion counts cash, marketable assets, paid development costs and contributed property, contributed before the first advance ^{[78] [79]}. A capital stack that relies on later equity or deferred fees may leave the loan in the 150 percent class ^[75].

The study and the appraisal echo each other. Read across from the guidelines' housing passage, market information can help an appraiser but does not serve as an evaluation ^[66], and an appraiser must make their own deductions for a property not yet stabilized ^[70]. Identical lease-up periods and rents in both documents, with no stated source in either, should prompt a question about which came first.

The market evidence is stale. The appraisal guidelines name changes in vacancy, absorption, rents, concessions and construction costs as signals that collateral should be looked at again ^[93]. A study dated before a supply wave or a rate move is a study of a different market.

The exception is assumed, not argued. A loan above the bank's limits needs a written justification that names every supporting credit factor ^[34], and it uses capacity in an exception budget the bank may already have spent ^[32].

None of these checks requires a study to decide the credit, the value or the bank's policy. Each asks only that the study's evidence be complete, sourced, dated and consistent with the rules the bank must apply. The program articles for EB-5 and SBA lending in this series run the same exercise against their own rulebooks.

Related Pages on Loan Analytics

• The Loan Analytics methodology series (all articles)
• How a feasibility study is done: the Loan Analytics methodology
• Bankable feasibility studies for conventional loans
• Loan data by state
• Loan data by asset class
• Worked examples (case studies)
• Worked Example: Conventional Bank Marina Acquisition and Dock Expansion Feasibility Study in Kentucky
• Worked Example: Conventional Bank Construction-to-Permanent Multifamily Feasibility Study in Nebraska
• Worked Example: Conventional Community Bank Construction-to-Permanent Short-Term Rental Feasibility Study in New York
• Feasibility Studies, Business Plans, and Investment Memos: What Banks Really Need
• SBA 7(a) Method: What 13 CFR 120 Tests and How a Feasibility Study Answers Each Test
• EB-5 Method: What USCIS Tests in the Business Plan and the Job Model, and How a Study Answers

Commission a Feasibility Study

Loan Analytics prepares lender-grade feasibility studies for SBA 7(a), SBA 504, USDA Business and Industry, conventional and EB-5 financing, built on the methods this series sets out. A study is not an appraisal: it tests whether a project's market, costs and cash flow can carry the debt a lender is asked to approve. To discuss a study for your project, contact Loan Analytics.

Daniel Smith, MAI, Loan Analytics