When a mini-perm lender will not refinance, who pays the balloon?

The dangerous misunderstanding around mini-perms is that construction completion somehow creates permanent financing. It does not. At the takeout date, lenders look afresh at the asset's income, value, occupancy and ability to service debt. If the new loan is smaller than the balloon payment, the owner must inject equity, sell, negotiate time or default. Cases from Manhattan, London, Hong Kong, South Africa and Bahrain show that the decisive question is not whether an asset was built, but whether there is a credible route to repay the existing lender.

Blackstone bought 1740 Broadway, a Manhattan office tower, for $605 million. It financed the purchase with a $308 million, ten-year interest-only mortgage.

Then L Brands left in 2022. L Brands was the retailer whose occupation of about 71% of the tower’s rentable area made it central to the building’s income. Its departure was not simply an awkward leasing problem. It removed the income stream on which a future lender would judge the building.

The loan entered special servicing, Blackstone stopped supporting the non-recourse asset, and the debt was later sold at a deep discount rather than refinanced conventionally.

The Manhattan office tower lost the retailer L Brands, whose departure removed a tenant representing about 71% of rentable area. Photo: Kidfly182 / Wikimedia Commons, CC BY 4.0.

That raises the question most borrowers ask too late: if a building is complete, why can it still run out of financing?

The finished-building illusion

People often say construction debt “converts” into permanent financing. It sounds reassuring. It is also usually incomplete.

A mini-perm is bank debt that carries an asset through construction and an early operating period. Practical Law describes it as financing normally replaced by permanent debt or project bonds shortly after operations begin. World Bank PPP guidance describes five-to-seven-year structures with escalating cash sweeps and, in hard versions, default if refinancing fails.

Completion may convert a construction loan into an operating-period loan. It does not make the balloon payment disappear.

A hard mini-perm says the borrower defaults if it cannot repay or refinance by the deadline. A soft mini-perm may avoid an immediate default, but applies rising interest margins, blocked dividends and cash sweeps, meaning spare cash is diverted to repaying debt.

Neither version obliges the original lender to write the next loan.

The interesting part is that this is not obscure fine print. It is the central bargain. The lender agreed to finance a defined period of risk, not an indefinite holding period.

The second underwriting nobody sees coming

A 2020 loan agreement filed with the US Securities and Exchange Commission makes the point unusually clearly. The supplied filing identifies its registrant through CIK 1680657, but the research brief does not provide the registrant’s name. The agreement itself matters because it makes conversion and a 12-month extension conditional rather than automatic.

The conditions included completion, a certificate of occupancy, a fresh appraisal, required loan-to-value and debt-service coverage, no default, and compliance with guarantor covenants. If the loan-to-value or debt-service coverage tests were not met, the borrower had to prepay principal. Even then, the extension was subject to the agent’s discretion and a fee.

That is what a takeout date really is: a second underwriting decision.

The original loan amount tells you what a lender believed during construction. The refinance amount tells you what a lender believes now.

The $28 million problem

Imagine an asset producing $10 million of annual net operating income. At a 5% cap rate, the yield used to value a property, it is worth $200 million. At a 6.25% cap rate, with exactly the same income, it is worth $160 million.

Nothing has broken in the building. Yet at 70% loan-to-value, available debt falls from $140 million to $112 million. That is a $28 million equity cheque simply to repay the original balance.

This is where most people stop looking. They see occupancy and assume safety. Lenders, however, lend against income, value and their own current standards, not against an owner’s memory of the original deal.

For US multifamily assets approaching stabilisation, Fannie Mae’s near-stabilisation execution permits refinancing from 75% physical occupancy, subject to certificates of occupancy for all residential units and expected underwritten stabilisation within four months. Its illustrative Tier 2 minimum debt-service coverage ratio, or DSCR, is 1.25x. In plain English, DSCR asks whether the cash coming in covers debt payments with room to spare.

US qualifying commercial real estate rules are tighter in some cases: 1.25x for multifamily, 1.50x for qualifying leased commercial real estate and 1.70x for other commercial real estate. HUD large-loan guidance can tighten market-rate refinancing above $75 million to approximately 75% loan-to-value and 1.30x DSCR.

Those are not universal global rules. They show the bigger point. A takeout lender has a fresh hurdle, and the old balance may not clear it.

The Federal Reserve estimated that almost $1 trillion, about 20% of outstanding US commercial real estate loans, matured during 2025 while borrowers faced tighter standards, lower values and rates above those prevailing at origination.

For a related look at how a balloon becomes an equity call, see The Door Marked ‘Refinancing’ Is Narrower Than It Looks.

London had a different answer

Canary Wharf Group, the owner of the major London business district and its mixed-use property portfolio, secured more than £2.1 billion of refinancing in 2024, including £610 million against retail assets. It extended its weighted-average debt maturity from 4.4 to 5.4 years.

But debt was only part of the answer. Brookfield, a global property investor, and Qatar Investment Authority, Qatar’s sovereign investment fund, are Canary Wharf Group’s shareholders. They supplied £400 million of additional capital.

The group also secured long lease extensions and broadened into residential, retail, hospitality and life sciences. It reported office occupancy of 88.2%, build-to-rent occupancy of 91.6% and retail occupancy of 97.4%.

At first glance, this looks like a story about a prestigious address. It is not. The lenders had evidence to underwrite: sponsor capital, leasing performance, a broader set of assets and a visible plan for repayment.

Put Canary Wharf beside 1740 Broadway and a pattern appears that neither case states outright. Refinancing is not a verdict on whether an asset is famous or finished. It is a verdict on whether the next lender can see a believable route back to its money.

More time is not less debt

New World Development, the Hong Kong property company, offers a different version of the same problem. After property-market weakness, high finance costs and valuation declines, it reported a HK$16.36 billion FY2025 loss. It obtained covenant waivers and completed a HK$88.2 billion loan refinancing and alignment.

That moved the immediate pressure. Short-term debt fell by roughly HK$35 billion to HK$6.6 billion. Yet total debt remained approximately HK$146.1 billion. The company withheld its final dividend and accelerated property sales and disposals.

Its refinancing eased near-term maturities but did not remove the wider debt burden. Photo: Tongzeit Mihoapci / Wikimedia Commons, CC BY-SA 4.0.

Here is the twist. A refinancing can solve a calendar problem without solving a balance-sheet problem. Lenders may grant time, but they often demand covenant alignment, dividend restraint, collateral and sale proceeds in return.

Rebosis Property Fund in South Africa faced the harsher version. It disclosed loan-to-value above 50%, interest cover below 2x, current liabilities exceeding current assets by R3.9 billion and reliance on lender roll-forwards or a proposed R6.3 billion property sale. The sale was expected to reduce loan-to-value to about 43%.

Its facilities were extended in short increments. Rebosis entered business rescue on 25 August 2022.

Short lender extensions could not substitute for the sale and balance-sheet repair needed to repay debt. Photo: Zak H / Pexels, Pexels licence (free commercial use).

The lesson is blunt. An extension is runway, not repayment.

When the pressure is deliberate

Not every mini-perm is a mistake. Al Dur 1, the independent water and power project in Bahrain, used an eight-year hard mini-perm with an approximately 80% balloon. Its terms included a 50-basis-point margin increase and a 100% cash sweep if refinancing did not occur on schedule.

The sponsors increased equity from about $300 million to $500 million.

That combination matters. The financing did not pretend the exit was automatic. It made the refinancing date visible, gave lenders economic pressure if it was missed, and provided more sponsor capital beneath the debt.

The evidence does not prove that every hard mini-perm is safe. It shows the opposite of the usual misconception: temporary debt can work when cash flows become dependable, sponsors can support the asset and a realistic takeout market exists. The mistake is not using temporary debt. The mistake is treating it as though it were already permanent.

What owners should do before the clock starts shouting

Start with the actual documents. Find the maturity date, balloon amount, appraisal conditions, extension fees, cash-sweep provisions, prepayment requirements and default triggers.

Then run a refinance model using lower value, weaker occupancy and tighter lending capacity than the original case. Do not ask only whether the business plan works. Ask whether the asset can repay the loan if the business plan works less well than expected.

Run several repayment routes at once: replacement debt, committed equity for the gap, an asset sale and an extension discussion before it becomes a last-minute plea.

Waiting for one route to fail before beginning the next is how borrowers lose bargaining power. For projects that need operating evidence before approaching capital providers, see Before you fund the factory, fund the evidence a factory needs.

What experienced investors ask first

Investors should be wary of the phrase “refinancing is planned”. Planned by whom? On what terms? Against what value? And who pays if the proceeds fall short?

The sharpest questions are simple. What debt can the asset support under current lending terms? What income supports that figure? Is one tenant, one sale or one lender extension carrying the entire plan? Who has legally committed the downside equity?

Blackstone’s decision at 1740 Broadway also carries an uncomfortable lesson. In a non-recourse structure, where the lender’s claim is generally limited to the collateral, surrendering an asset can be economically rational if new equity would not recover its value. That does not make it painless. It means the documents allocated the downside before the trouble arrived.

GI Network would begin a mini-perm mandate by mapping each repayment route against the maturity timetable: refinance proceeds under current underwriting, sponsor equity under downside values, sale proceeds after timing risk, and extension conditions under the existing documents. We would test whether cash flows can genuinely support the proposed takeout, align materials with replacement lenders’ requirements, map the appropriate capital providers and rehearse the objections an investment committee is likely to raise. For a wider guide to raising money against evidence rather than optimism, read How to raise capital for a solar project: stop selling panels, sell certainty.

GI Network's view: Treat a mini-perm as a debt that must be repaid four different ways on paper before it is repaid once in real life.

The Four-Route Repayment Test

Before accepting a mini-perm, put the proposed balloon through the Four-Route Repayment Test:

1. 1.Refinance: can current cash flow and current value support enough new debt?
2. 2.Recapitalise: is there a credible sponsor or investor willing to fund the gap?
3. 3.Realise: can an asset sale repay the balance within the available time?
4. 4.Reschedule: will the existing lender extend, and what cash, covenants or control will it demand?

If only one route works, the asset has a hope, not a repayment plan. If none works under a modest downside case, the mini-perm is not buying time. It is merely naming the date when the problem becomes unavoidable.