Refinancing Risk When Property Values Fall

When a property is worth less than its loan, a replacement lender sizes the new loan off current value and income, and the borrower repays the difference with new cash. That cash is larger than the negative equity because the new loan is capped below full value, and closing costs and reserves add to the total. If the capital cannot be found, the paths left are an extension that only delays the problem, or a sale or transfer of the property to the lender through foreclosure or deed in lieu.

Why the refinancing gap is larger than the negative equity

Negative equity is the amount by which the existing loan exceeds property value. The refinancing gap is the existing payoff, plus closing costs and required reserves, minus whatever a new lender will advance. The two measures diverge because a new lender never advances the full value of the property.

An illustrative case shows the effect. A loan of $44 million sits on a property now worth $35 million. A replacement loan of $25 million would require $19 million of equity to close. The equity check is far larger than the shortfall between loan and value, because the new loan is sized to the lower value and the old balance still has to be retired in full.

Value falls through the capitalization rate. A cap rate is one year of net operating income, or NOI, divided by price, so with NOI held constant a higher cap rate produces a lower value. When investors demand higher income returns from private real estate to justify its risk premium over Treasuries and other low-risk fixed-income instruments, pricing moves through exactly this channel.

The payoff itself does not move when value falls. Only the replacement proceeds move.

Which sizing test sets the new loan amount

A replacement lender runs several tests and funds the smallest result. Loan-to-value divides the loan by property value. Debt service coverage ratio, or DSCR, divides underwritten NOI by scheduled debt service. Debt yield divides underwritten NOI by the loan balance. Every threshold varies by lender and loan type, so any number used in an analysis belongs to that deal’s term sheet and carries no universal weight.

Each test converts into a maximum loan. The value test gives the loan-to-value cap multiplied by value. The coverage test gives NOI divided by the required coverage, divided again by the annual debt constant. The debt yield test gives NOI divided by the minimum debt yield. The borrower receives the lowest of the three figures, and the test that produces it is the one worth diagnosing. A $25 million new loan against a $35 million value, as in the illustration above, shows a lender advancing well under full value.

The debt constant is annual scheduled payment divided by loan balance. With amortization it exceeds the interest rate, since principal is inside the payment. A higher constant shrinks the coverage-sized loan at unchanged NOI, so a rate move or an amortization change can cut proceeds without any change in value.

The NOI used in the numerator feeds value and coverage at the same time. Income that is overstated inflates the appraised value and the coverage ceiling together. A loan sized on in-place income also cannot credit income that has not yet arrived, which is why a balance coming due before lease-up finishes leaves a gap even when projected income would support the debt.

Worked example: refinancing a $40 million loan as the cap rate moves from 7.5% to 8%

This hypothetical compares two cases on the same basis: a $40 million payoff, annual NOI of $2.4 million, closing costs of $0.8 million, required reserves of $1.2 million and no spare property cash. The replacement lender applies a 65% maximum loan-to-value and a 1.25 times minimum coverage. A 10% minimum debt yield is a further test. Scheduled payments equal 8% of the loan balance each year, a debt constant that bundles interest and principal. The base case values the property at a 7.5% cap rate. The stressed case uses an 8% cap rate with every other input unchanged.

Base case value is $2.4 million divided by 0.075, which equals $32 million. Negative equity is $40 million minus $32 million, or $8 million. The existing loan-to-value is $40 million divided by $32 million, or 125%.

The value test allows $32 million multiplied by 0.65, or $20.8 million. Maximum annual debt service is $2.4 million divided by 1.25, or $1.92 million, and dividing that by the 8% constant gives a coverage ceiling of $24 million. The debt yield test gives $2.4 million divided by 0.10, also $24 million. Value sets the loan at $20.8 million.

The principal paydown is $40 million minus $20.8 million, or $19.2 million. It has two parts: the $8 million of negative equity and a cushion of $11.2 million, which is $32 million multiplied by the 35% of value the lender does not advance. Closing costs and reserves total $2 million. New capital is $19.2 million plus $2 million, or $21.2 million. Debt service on the new loan is $20.8 million multiplied by 0.08, or $1.664 million.

At an 8% cap rate, value is $2.4 million divided by 0.08, or $30 million, a decline of $2 million or 6.25%. Negative equity is $10 million. The value test allows $30 million multiplied by 0.65, or $19.5 million, while the coverage and debt yield ceilings stay at $24 million, so value still sets the loan.

The stressed paydown is $40 million minus $19.5 million, or $20.5 million, made up of $10 million of negative equity and a $10.5 million cushion. New capital is $20.5 million plus $2 million, or $22.5 million. Debt service is $19.5 million multiplied by 0.08, or $1.56 million.

A 6.25% fall in value raises the capital call from $21.2 million to $22.5 million, an increase of $1.3 million. That equals the $2 million value decline multiplied by 0.65, because every dollar of lost value removes 65% of that dollar from proceeds. Debt service falls from $1.664 million to $1.56 million, so the smaller loan costs less to carry while requiring more cash to close.

A sale on the same basis uses $1 million of selling costs. Base net proceeds are $32 million minus $1 million, or $31 million, which leaves $9 million of the $40 million payoff unpaid. Stressed net proceeds are $30 million minus $1 million, or $29 million, leaving $11 million unpaid.

Two reference points frame the range. The cap rate at which value equals the loan is $2.4 million divided by $40 million, or 6%. At that rate the value test would allow $26 million, but the $24 million coverage and debt yield ceilings set the loan, so the paydown is still $16 million and new capital is $18 million. The value test sets the loan whenever the cap rate exceeds 6.5%, which is the 65% maximum loan-to-value multiplied by the 10% debt yield.

Why a stabilizing market does not close a single maturity gap

The refinancing queue is large. About 20% of outstanding CRE loans, just shy of $1 trillion, were forecast to mature in a single year, and some of those borrowers had not yet secured refinancing.

Transaction-based price indexes can stabilize while the maturity pressure remains. Stabilizing prices and fundamentals sit alongside a continuing vulnerability from upcoming refinancing needs, and distressed sales stay possible for borrowers who cannot replace their loans.

Core open-ended funds sold properties at nearly 6.5% cap rates in one quarter and at 5.9% in the next, a sign that private transaction valuations had corrected. That range describes the properties those funds sold. A particular building with weaker income or a thinner market is valued at its own cap rate, and its replacement loan is sized from that value.

The gap therefore gets measured one asset at a time. A market-level recovery in pricing changes the odds for the sector. The arithmetic for a single loan depends on that loan’s NOI, its appraised value and the term sheet in hand.

“CRE markets showed some signs of stabilizing prices and fundamentals, although the potential for distressed commercial property sales remains if CRE borrowers who need to refinance their mortgages are unable to do so.”

How loss models treat loans above a stressed 0.8 loan-to-value

Scenario models of insurer mortgage portfolios draw a hard line. Mortgages with a stressed loan-to-value of 0.8 or higher are assumed to default, and loss given default is set at 30% of the mortgage balance.

Such a scenario also assumes all loans default, including loans that do not need refinancing, and it cites evidence of strategic renegotiation by borrowers of CMBS and bank loans. The threshold and the loss rate are modeling choices for a portfolio. They are not a forecast for any individual loan.

The 30% figure is a share of balance, so it fixes a loss severity for the model without measuring the gap between payoff and actual recovery on a given property. In the worked example, the base case already sits at a 125% loan-to-value, far above that threshold.

Duration of the income decline matters to the same models. Lasting downward changes in CRE cash flows cause lasting valuation declines and eventual losses for holders, while gradual losses are unlikely to cause large disruptions across the financial system. A temporary dip in NOI can be bridged. A permanent one resets the value, and the cap rate math in the example then applies.

What happens when the gap cannot be funded

Four paths are open: an extension, a recapitalization with new capital, a sale, or a transfer of the property through foreclosure or deed in lieu. Some operators who are underwater with no prospect of lower rates end up selling at whatever price clears or surrendering the property to the lender.

An extension changes timing only. With the same NOI and the same cap rate, the base case paydown is still $19.2 million at the new maturity. Only higher NOI, a lower cap rate or a different debt constant moves that figure, and carry cash spent while waiting adds to the requirement.

A recapitalization and a sale answer different questions. In the base case, $21.2 million of new capital retains an asset valued at $32 million with a $20.8 million loan. A sale at the same value ends with $9 million of the payoff unpaid, and $11 million unpaid in the stressed case, which requires another payment source or an agreed reduction in the debt.

Resolution can raise the payoff. A real estate debt agreement may provide for a higher interest rate after default or after transfer of the property as collateral, so the amount due on the day of a workout can exceed the scheduled balance.

Pricing opens for buyers with capital once liquidity leaves the market, which is the other side of the same transaction.

Concentration compounds each path. A portfolio concentrated in one local property market faces the same value decline on every asset at once, so proceeds that one asset was expected to contribute toward another asset’s refinancing shrink together.

“Basically, real estate gets interesting when the liquidity gets taken out of it.”

Pitfalls that understate the refinancing gap

Equating negative equity with the capital call is the most common arithmetic slip. In the base case negative equity is $8 million, yet new capital is $21.2 million, because the lender withholds 35% of value as cushion and costs and reserves add $2 million. A model that stops at negative equity understates the check by more than half.

Testing only debt service coverage skips the binding constraint. In both example cases the coverage ceiling is $24 million, well above the value-based loan of $20.8 million or $19.5 million. A property can pass coverage comfortably and still face a paydown of $19.2 million or more, since the three tests are independent and the lowest controls.

A cap rate is one year of NOI divided by price. It measures an annual property income yield, not a holding period return or IRR, so holding period is not an input to direct capitalization. Moving from 7.5% to 8% at constant NOI cuts value by $2 million, and that change flows through the loan-to-value test at 65% no matter how long the asset is held.

Using market pricing signals in place of the asset’s own valuation produces a gap estimate that does not apply. A sector-level move from nearly 6.5% to 5.9% in sale cap rates says little about a building appraised at 7.5%, and a stabilizing price index leaves each borrower’s maturity gap untouched.

Counting one pool of cash against several properties understates aggregate funding needs. When valuations fall across a concentrated portfolio, each asset requires its own paydown, and the same liquidity or the same expected refinancing proceeds cannot cover all of them.

Questions that expose a refinancing gap before maturity

What is the full payoff on the anticipated closing date? Accrued interest and fees can place the payoff above the stated principal, and default-rate amounts can add to it.

Which sizing test produces the smallest loan? Each of the three ceilings needs its own calculation before one is called binding.

What debt constant does the new lender assume? Amortization raises the constant above the interest rate and cuts the coverage-based ceiling.

Which NOI figure feeds the appraisal and the coverage test? A single overstated income line inflates both value and proceeds.

At what cap rate does value equal the existing loan? In the example, $2.4 million divided by $40 million gives 6%, and a gap remains at that rate.

How much new capital does each step of cap rate expansion add? In the example, a move from 7.5% to 8% added $1.3 million to the capital call.

What cash remains after costs and reserves? Cash posted to a reserve cannot also retire principal.

Does an extension include an operating milestone? A later maturity with unchanged NOI and cap rate leaves the same paydown and adds carrying cost.

Which other holdings depend on the same cash or the same market? Concentration in one local market means the decline reaches several assets together.