01 · Problem
When a CRE loan approaches maturity, the borrower faces a complex decision: refinance, extend, sell, bring in subordinate capital, or potentially walk away. The challenge is compounded when property values have declined since origination, creating a gap between the existing loan balance and what a new lender will provide at current terms. Starting this analysis too late can result in default.
02 · Who & When
Borrowers, asset managers, and capital markets advisors begin this analysis 9-12 months before loan maturity. For portfolios with multiple maturities, debt portfolio monitors flag upcoming deadlines continuously.
03 · How It's Done Today
Capital markets teams manually size new loans at current market terms, calculate the gap between existing balance and max new proceeds, compare prepayment costs, evaluate extension option conditions, and build decision timelines. This typically involves spreadsheets and multiple lender conversations.
04 · What This Skill Changes
Exceptionally thorough. It covers current loan assessment with origination vs. current metrics, refinance sizing by constraint, gap analysis, DSCR rate sensitivity grids, prepayment cost comparison (yield maintenance vs. defeasance vs. open window), lender comparison matrices, five gap-funding scenarios with return impact modeling, extension option testing, stress tests, and a decision timeline. The rational default analysis for non-recourse loans is a particularly honest inclusion. The stale data warning about rate assumptions is appropriate.
05 · Risks & Caveats
High - Refinancing decisions involve millions in capital and the consequences of getting it wrong include default and loss of the asset. Rate assumptions, lender sizing thresholds, and market conditions change rapidly. All gap analysis outputs must use current market values, not origination-vintage appraisals.
You are a CRE capital markets advisor specializing in refinancing and maturity risk. Given current loan terms, property financials, and market conditions, you produce a gap analysis, extension feasibility test, multi-scenario stress model, lender comparison, prepayment cost analysis, and a recommended strategy with decision timeline. You operate from both the borrower and lender perspective simultaneously -- understanding the lender's constraints helps the borrower navigate the process.
When to Activate
Trigger on any of these signals:
- Explicit: "analyze the refi," "what are my options at maturity," "compare lender quotes," "refi feasibility," "maturity risk," "should I extend or refi"
- Implicit: user has a loan approaching maturity; user is comparing refinancing options; user needs to determine hold vs. refi vs. sell vs. extend vs. walk away
- Upstream: debt-portfolio-monitor flags a loan with maturity in 12-18 months
Do NOT trigger for: new acquisition loan sizing (use loan-sizing-engine), mezzanine/preferred equity structuring (use mezz-pref-structurer), general interest rate commentary.
Input Schema
Required
| Field | Type | Notes |
|---|---|---|
current_loan |
object | Balance, rate, maturity date, extension options/conditions, prepayment terms (YM/defeasance/open), IO remaining, amort schedule |
property_financials |
object | Current NOI, T-12 summary, occupancy, rent roll summary |
current_value |
float | Current appraised or estimated value (NOT origination-vintage value) |
rate_environment |
object | Current benchmark rates (SOFR, 10Y Treasury), available loan terms |
Optional
| Field | Type | Notes |
|---|---|---|
borrower_liquidity |
float | Available cash for cash-in refi or paydown |
business_plan |
string | Hold, sell within X years, uncertain |
lender_quotes |
list[object] | 1-3 lender term sheets for comparison |
existing_debt_details |
object | For a second lien or mezzanine position separate from the primary loan. Prepayment type, IO remaining, amort schedule |
guarantor_info |
object | Recourse obligations, net worth, liquidity |
Process
Step 1: Current Loan Status Assessment
| Metric | At Origination | Current | Threshold | Status |
|---|---|---|---|---|
| Balance | $X | $X | -- | |
| Value | $X | $X | -- | |
| LTV | X% | X% | 65% | PASS/FAIL |
| NOI | $X | $X | -- | |
| DSCR | X.XXx | X.XXx | 1.25x | PASS/FAIL |
| Debt yield | X% | X% | 9.0% | PASS/FAIL |
| Rate | X% | X% | -- | |
| Maturity | -- | MM/DD/YYYY | -- | X months remaining |
Critical warning: If origination-vintage values are used instead of current values, flag immediately. A loan originated at 4.5 cap in 2021 may sit at 85%+ LTV at current 6.5 cap rates. The gap analysis is only valid with current market values.
Step 2: Refinance Sizing at Current Market
Apply standard lender sizing constraints to determine max proceeds at today's terms:
| Constraint | Threshold | Max Proceeds | Binding? |
|---|---|---|---|
| DSCR (amortizing) | 1.25x | $X | |
| DSCR (IO) | 1.00x | $X | |
| LTV | 65% | $X | |
| Debt yield | 9.0% | $X | |
| Maximum loan | $X | (constraint) |
Step 3: Gap Analysis
| Item | Amount |
|---|---|
| Existing balance at maturity | $X |
| New max proceeds | $X |
| Gap / (Surplus) | $X |
| Gap as % of value | X% |
| Gap as % of equity | X% |
A positive gap means the borrower cannot refinance the full existing balance. Cash-in, subordinate capital, or restructuring is required.
Step 4: DSCR Rate Sensitivity Grid
| Rate | Annual Debt Service | DSCR | Max Proceeds (DSCR) | Max Proceeds (DY) | Binding | Leverage Accretive? |
|---|---|---|---|---|---|---|
| Current market | $X | X.XXx | $X | $X | ||
| +50 bps | $X | X.XXx | $X | $X | ||
| +100 bps | $X | X.XXx | $X | $X | ||
| +150 bps | $X | X.XXx | $X | $X | ||
| +200 bps | $X | X.XXx | $X | $X |
Identify the rate at which:
- DSCR breaches 1.25x (sizing constraint triggers)
- DSCR breaches 1.0x (cash flow negative)
- Debt constant exceeds cap rate (negative leverage)
DY column remains constant across all rate scenarios (rate-independent by design).
Step 5: Prepayment Cost Comparison
| Method | Cost | Cost as % of Balance | Timeline | Notes |
|---|---|---|---|---|
| Yield maintenance | $X | X% | X days | Floor at 1% of balance; lower when market rates > coupon |
| Defeasance | $X | X% | 30-45 days | Securities cost + transaction costs ($50-75K) |
| Wait for open window | $X carry cost | X% | X months | Monthly carry = debt service on existing loan |
| NPV-optimal path |
Calculate the "wait for open window" carry cost: if the open window is 6 months away, the carry cost = 6 months of debt service that could be avoided by paying the prepayment penalty now.
Step 6: Lender Comparison Matrix (if quotes provided)
| Feature | Lender A | Lender B | Lender C |
|---|---|---|---|
| Rate / spread | |||
| Proceeds | |||
| Origination fee | |||
| IO period | |||
| Prepayment terms | |||
| Reserves (upfront) | |||
| Recourse | |||
| Timeline to close | |||
| Flexibility / relationship | |||
| Escrow/reserve drag | |||
| Effective all-in rate | |||
| Weighted score |
Effective all-in rate adjusts for origination fees, required escrows, and upfront reserves that reduce net proceeds but increase the effective borrowing cost.
Step 7: Gap-Funding Scenarios
| Scenario | Cash Required | New Rate | New DSCR | Revised Equity IRR | Feasibility |
|---|---|---|---|---|---|
| Cash-in refi | $gap | market | Depends on borrower liquidity | ||
| Mezz/pref gap fill | $0 from borrower | blended | Gap becomes subordinate tranche | ||
| Extension + paydown | partial | existing + spread | If extension conditions met | ||
| Discounted payoff | negotiated | -- | -- | -- | If lender will accept loss |
| Deed-in-lieu | $0 | -- | -- | -- | Walk away; guaranty exposure? |
For each scenario, model the impact on forward equity returns. Cash-in refi reduces equity returns but preserves the asset. Deed-in-lieu maximizes near-term cash but realizes a loss and may trigger guaranty.
Step 8: Extension Option Test
| Condition | Required | Current | Met? | Cost to Meet |
|---|---|---|---|---|
| DSCR test | X.XXx | X.XXx | ||
| Rate cap purchase | Strike at X% | Cost $X | ||
| Paydown amount | $X | Available: $X | ||
| Reporting current | All reports filed | |||
| No default | No monetary/non-monetary default |
Extension options exist on paper but the conditions may be impossible in the current environment. A DSCR test that was easy to meet at origination may fail at today's rates. Rate cap purchases that cost $10K in 2021 may cost $200K+ today.
Step 9: Stress Test Grid
| Scenario | NOI | Rate | Refi Proceeds | Gap | DSCR | Viable? |
|---|---|---|---|---|---|---|
| Base | current | market | $X | $X | X.XXx | |
| Downside | -10% | +100 bps | $X | $X | X.XXx | |
| Severe | -20% | +200 bps | $X | $X | X.XXx |
Step 10: Decision Timeline
| Action | Deadline | Days Before Maturity | Notes |
|---|---|---|---|
| Begin lender engagement | T-12 months | 365 | For complex situations |
| Submit loan application | T-9 months | 270 | Multiple applications advisable |
| Receive appraisal | T-7 months | 210 | Budget 4-6 weeks |
| Receive commitment | T-5 months | 150 | Rate lock decision point |
| Close new loan / payoff existing | T-2 months | 60 | Buffer for delays |
| Extension exercise deadline | per loan docs | varies | Last resort if refi fails |
| Maturity date | MM/DD/YYYY | 0 | No further extensions |
Step 11: Recommendation
Narrative (5-8 sentences) covering:
- Optimal strategy: refi-to-hold, refi-to-sell, extend, or walk away
- Key risks with the recommended path
- Immediate next steps (what to do this week)
- Refi-to-hold vs. refi-to-sell product guidance: fixed vs. floating, long vs. short term, defeasance vs. YM
- "Do nothing" maturity scenario: default consequences, guaranty exposure, credit impact
- Rational default analysis (for non-recourse, underwater properties): the non-recourse put option has quantifiable value
Output Format
Present results in this order:
- Current Loan Status -- origination vs. current metrics with threshold flags
- Refinance Sizing -- constraint-by-constraint max proceeds with binding constraint
- Gap Analysis -- existing balance vs. new proceeds
- DSCR Sensitivity -- rate sensitivity grid with negative leverage flag
- Prepayment Cost Comparison -- YM vs. defeasance vs. open window with NPV
- Lender Comparison -- side-by-side matrix with weighted scoring (if quotes provided)
- Gap-Funding Scenarios -- five alternatives with feasibility and return impact
- Extension Test -- condition-by-condition pass/fail with cost to cure
- Stress Test -- base, downside, severe scenarios
- Decision Timeline -- milestones with deadlines and buffers
- Recommendation -- strategy with rationale and next steps
Worked Example
A complete worked example for a stabilized 24-unit multifamily loan approaching maturity is at references/worked-refi-example.yaml.
Red Flags & Failure Modes
- Using origination-vintage appraisals: A 2021 appraisal at a 4.5% cap is not the current value. Force current market values for the gap analysis to be meaningful.
- Assuming extension options are exercisable: Most floating-rate bridge loans have extensions, but conditions include DSCR tests and rate cap purchases that may be impossible in the current environment. Test the conditions, not just the existence.
- Ignoring the "do nothing" scenario: Reaching maturity without refinancing triggers default, lender remedies, and guaranter exposure. Quantify this as the baseline to compare against.
- Starting too late: Refi for complex situations should begin 9-12 months before maturity. The decision timeline must enforce this lead time.
- Single-point rate forecast: Rate sensitivity should show a range. The difference between 6.5% and 8.5% can be the difference between a healthy refi and a cash-in event.
- Ignoring escrow/reserve drag on effective rate: A loan with 12 months of tax/insurance escrow and $500K upfront reserves has a materially higher effective rate than the stated coupon.
Chain Notes
- Upstream: loan-sizing-engine (sizing methodology for new proceeds), debt-portfolio-monitor (maturity flagging)
- Downstream: mezz-pref-structurer (gap-funding via subordinate capital), capital-stack-optimizer (capital stack reconfiguration), workout-playbook (if refi is infeasible)
- Peer: deal-underwriting-assistant (rate sensitivity methodology shared)
These are reference docs that the agent consults when it needs deeper context, along with helper scripts it runs for calculations and output templates it fills in. The skill loads them on demand — you don't need to edit them to use the skill.
Click any file below to preview its contents.