The Reserve That Isn’t Excess: How Lenders Size and Control CapEx, TILC, and Debt Service Reserves—and Why Release Tests Quietly Dictate Real Leverage

The Reserve Account as Embedded Lender Control

Reserve accounts on a loan term sheet look straightforward: capital set aside for capital expenditures, tenant improvement and leasing commissions (TILC), and debt service. They appear fungible, compartmentalized, and released on mechanical criteria. In practice, reserve sizing, release sequencing, and subordination hierarchy function as a second layer of leverage control, one that survives covenant compression and operates independently of traditional loan-to-value or debt-service-coverage calculations.

The structural insight is simple: a reserve is not capital. It is a lender claim on future cash flow, held in escrow, released only when borrower performance satisfies conditions the lender designed and can modify through amendment. Sizing, subordination, and release mechanics transform reserves from safety buffers into instruments of capital control.

CapEx Reserves: Sizing as Constraint, Release as Discretion

CapEx reserves typically range from 1 to 5 percent of the loan amount, depending on asset class, vintage, and sponsor credit. The lender calculates this as a fraction of the loan, not as a true forecast of capital need. The result is that reserve sizing is often disconnected from actual capital requirements.

Consider a stabilized office or industrial asset with deferred maintenance. The lender underwriting model may assume $150,000 in annual CapEx, but the reserve is sized at 2 percent of the $50 million loan, or $1 million. Over the first three years, that reserve exhausts in 6 to 7 years of actual spend. But the release test is not a schedule. It is conditional: CapEx reserves release only when debt service coverage ratio (DSCR) exceeds a threshold (typically 1.25x to 1.35x), and often only when certifications from the borrower's engineer confirm that capital has been deployed and capitalized.

This creates a paradox. In years when cash flow is strong, DSCR is high, and release tests pass, the borrower may not need the reserve. In years when cash flow is stressed, DSCR falls, and release tests fail, the borrower cannot access capital and cannot fund necessary maintenance. The reserve, designed to protect the lender, often prevents the borrower from preserving asset condition.

Subordination deepens the constraint. Many CapEx reserves are junior to debt service reserves. The borrower cannot access CapEx reserves until the debt service reserve is full. In a stressed year, the debt service reserve must be replenished before CapEx reserves are released, compressing borrower flexibility further.

TILC Reserves: Renewal Optionality Disguised as Contingency

TILC reserves are sized to fund tenant improvement costs and leasing commissions for lease renewals and new leases. Sizing is typically a per-unit or per-square-foot assumption applied to anticipated turnover. A 50,000-square-foot office building with 15 percent annual turnover and $30 per-square-foot TI plus 6 percent of rent in commissions might require $250,000 to $350,000 annually in TILC spend.

But lenders often size TILC reserves at a fraction of this. A $5 million TILC reserve on a $50 million loan is 10 percent of the loan amount. If annual TILC spend is $300,000, that reserve lasts 16 years. But it does not. Most loan documents require TILC reserves to be replenished if they fall below a trigger (typically 6 to 12 months of projected spend). Replenishment comes from cash flow, and replenishment is itself subject to a release test.

The result is that TILC reserves function as a constraint on renewal economics. If the borrower wants to offer higher TI or higher commissions to retain quality tenants, the reserve may not support it. The lender controls the terms of renewal through the reserve ceiling and release mechanics. The borrower cannot fund above-market TI without the lender's explicit consent, which often comes in the form of a covenant waiver or amendment.

This is particularly powerful in competitive leasing markets. Borrowers with strong credit and high-quality assets can sometimes negotiate TILC reserve release thresholds that allow flexibility. Sponsors with marginal credit or transitional assets find TILC reserves become binding constraints on leasing strategy.

Debt Service Reserves: The Hidden Leverage Multiplier

Debt service reserves (DSRs) are the lender's true control lever. DSR sizing is typically 6 to 12 months of debt service. On a $50 million loan at 5 percent interest, with a 25-year amortization, monthly debt service is roughly $290,000. A 6-month DSR is $1.74 million. A 12-month DSR is $3.48 million.

The DSR sits in escrow, subordinate to nothing (it is senior to all other reserves), and is released only when the borrower maintains DSCR above a threshold for a specified period (typically 1.20x to 1.30x DSCR for two or more consecutive quarters). Once released, the DSR cannot be re-funded without another amendment or consent.

The structural consequence is that the DSR functions as a hidden leverage multiplier. A borrower with a $50 million loan and a 12-month DSR of $3.48 million has effectively $53.48 million of debt service obligations to meet before cash is available for distributions. The DSR is financed by the loan (it is part of the loan proceeds), but it is not distributed to the sponsor. It remains in escrow, compressing available cash by 7 percent of the loan amount.

Release tests on DSRs are often asymmetric. The borrower must achieve 1.25x DSCR for two consecutive quarters to trigger release. But if DSCR falls below 1.15x, the lender may require the DSR to be re-funded from cash flow. This creates a ratchet: the DSR is easy to trigger (two quarters of good performance), but difficult to stay released (any stress can trigger a re-funding obligation).

Many loan documents also subordinate other reserves to the DSR. CapEx reserves, TILC reserves, and even capital improvement reserves are junior. The borrower cannot access them until the DSR is fully funded and released. This creates a waterfall: debt service, then DSR funding, then other reserve replenishment, then CapEx drawdown, then cash available for distributions.

Release Tests as Real Leverage Constraints

The critical insight is that release tests quietly dictate real leverage. A borrower with a 75 percent loan-to-value (LTV) ratio and a 1.25x DSCR may appear to have 25 percent equity cushion. But if the DSR is 12 months of debt service and CapEx and TILC reserves total another 5 percent of the loan, the borrower's true cash-available-for-distribution is compressed by 17 percent of the loan amount. The effective leverage is not 75 percent LTV; it is closer to 90 percent when reserve holdbacks are modeled as embedded debt service obligations.

Release tests amplify this. If DSCR must exceed 1.25x for two consecutive quarters to release reserves, the borrower must achieve a 1.25x coverage ratio before any reserve capital is available. In stressed scenarios, this means the borrower cannot access capital reserves precisely when they are most needed.

The lender's optionality is built in. If market conditions deteriorate, the lender can tighten release tests through amendment, requiring higher DSCR thresholds or longer performance periods before release. If the borrower needs capital, the lender can negotiate a waiver or release, extracting additional fees or amended terms in exchange. The reserve becomes a negotiation tool, not a mechanical safety valve.

This is path-dependent. Early in the cycle, when spreads are tight and competition is fierce, lenders size reserves conservatively and release tests generously. As credit cycles compress and delinquencies rise, lenders tighten reserve sizing, raise release thresholds, and subordinate reserves more aggressively to debt service. Borrowers who financed at cycle peak find themselves unable to access reserves designed to protect them, because release tests were calibrated to cycle trough stress.

Why This Matters for Borrower Economics

The structural consequence is that reserve mechanics function as a second layer of leverage control, independent of covenants and loan-to-value ratios. A borrower can have a strong DSCR and still be unable to fund capital or leasing if reserve release tests are not satisfied. A borrower can have low LTV and still be effectively over-leveraged if reserves are large, subordinated, and release tests are tight.

This is not a flaw in loan design. It is intentional. Lenders use reserves to transfer optionality from borrower to lender. The borrower gets the capital upfront (it is funded at closing), but cannot access it without the lender's permission. The lender gets to decide when reserves are released, can modify release tests through amendment, and can use reserve access as a negotiation lever in stressed scenarios.

For sponsors, the implication is that reserve mechanics must be negotiated as carefully as covenants and loan-to-value ratios. Reserve sizing should be tied to true capital forecasts, not arbitrary percentages of the loan. Release tests should be mechanical, not discretionary, and should not require lender consent. Subordination hierarchies should be transparent, and borrowers should have clarity on the order of priority before closing.

For lenders, reserve mechanics are a tool for managing tail risk and preserving optionality. But they are also a constraint on borrower flexibility and a source of friction in stressed scenarios. Lenders who size reserves aggressively and tighten release tests in downturns often find that borrowers cannot execute capital strategies that would preserve asset value, leading to worse outcomes for both parties.

What I'm watching next

• Reserve release language in new loan documents, particularly thresholds for DSCR and duration of performance periods, as a leading indicator of lender appetite for borrower flexibility in a tightening cycle.
• Amendment activity around reserve access in stabilized portfolios, which often signals whether borrowers are hitting release test constraints and lenders are using reserve access as a renegotiation lever.
• Sponsor-side pushback on reserve subordination in new deals, particularly in competitive markets, where borrowers have leverage to demand mechanical release tests and transparent reserve hierarchies.