It's Tuesday 4 PM. You call your main bank: you need a $15M additional line to close an unexpected gap this quarter. The friendly executive who's known you for 5 years tells you your request requires "committee analysis" and you'll have an answer in 3-4 weeks. Three weeks. You have 10 days.
That's the moment where you find out whether you built banking relationships in calm or improvised them under pressure. Credit you get under pressure is 3-5x more expensive and 60% slower than credit you get in a good moment. The experienced CFO knows: you ask for the line BEFORE you need it.
This module is about building a diversified banking portfolio with committed lines, negotiated covenants, and enough cushion to survive any adverse scenario without asking for favors. It's not about minimizing financial cost — it's about minimizing fragility.
Let's go.
Andina has 4 banks: $60M with the main one, $40M secondary, $30M state, $25M investment bank. Total: $155M of lines. If EBITDA drops 35% in a recession, how much of that total line is REALLY available?
In plain language
Before the mechanics, the four basic questions.
Why do this at all?
Because liquidity under stress is what separates companies that survive a recession from those that go into technical default with positive P&L. A line you assume to have, but the bank will freeze when EBITDA falls, is NOT liquidity — it's an accounting illusion. Building real banking relationships means ensuring the line is there when you need it, not when the credit committee tells you 3 weeks later.
Who manages them?
CFO and treasurer share ownership. CFO maintains the relationship with banking relationship partners (commercial manager, in some cases regional director). Treasurer operates day-to-day: drawdown, repayments, covenant compliance, monthly reports. Both negotiate renewals and new facilities together. People who should NOT manage banking relationships: operations directors, AR managers, controllers.
When does it show up?
Continuously: quarterly review of line usage. Annually: line renewal and covenant renegotiation. On events: M&A, large capex, ownership change, refinancing. And critically: 12-18 months BEFORE any anticipated moment of need. If you know in 18 months you'll do a large acquisition, starting to talk to banks NOW is the difference between getting competitive financing and accepting whatever they offer.
What if we neglect them?
Four concrete consequences: (1) Concentration in 1-2 banks: when one of them changes policy or gets difficult, you lose 50%+ of your capacity. (2) Uncommitted lines that look available but the bank can unilaterally cancel. (3) Tight covenants negotiated under pressure from lack of alternatives, limiting your operational decisions for years. (4) Recession surprise: you discover the "$100M line" becomes $30M usable when banks tighten. And by then it's too late to diversify.
Andina S.A. — the relationship portfolio
Andina has four active banking relationships, built over 12 years. The strategy behind the portfolio is deliberate, not accidental:
Main Bank — 12-year relationship, $60M committed line (the largest). It's the core relationship: your first call when you need something, the bank that knows you best, the one renewing terms with least friction. Covenant: Net debt/EBITDA <3.5x (comfortable).
Secondary Bank — 8-year relationship, $40M committed line. Serves two purposes: (a) operational backup so you don't depend 100% on the main, (b) healthy competition — when you renegotiate with the main, having the secondary with an alternative proposal lowers the price. Covenant: Net debt/EBITDA <3.0x (tight).
State Bank — 4-year relationship, $30M committed line, $0 drawn. Government bank with subsidized rates, more bureaucratic but MORE PATIENT in recession (doesn't freeze lines as fast as private). More generous covenant: 4.0x. Strategic for liquidity under stress.
Investment Bank — 2-year relationship, $25M committed, 100% drawn. It's the bridge facility: highest rate, strictest covenants, but gave you capacity when traditionals were slow. Covenant: 2.5x (very tight). Useful but NOT reliable under stress — first bank to cut you when EBITDA falls.
Total committed line: $155M. Current drawn: $80M. Nominal available: $75M. BUT under modeled stress of −30% EBITDA, that available drops to ~$40-50M because the investment bank freezes and the secondary reduces. That is the honest metric.
The visual below lets you play with stress and see how the four relationships behave.
Portfolio, live
Four cards, one per bank, with its line, drawn, effective availability, covenant cushion, and health badge.
The critical experiment: push stress to −30%. Watch the investment bank fall to "at risk" immediately (covenant 2.5x breaches), but the state and main banks stay "healthy". At −50% even the main enters the "watch" zone. Total "available under stress" falls materially.
Interactive visual
Bank relationship portfolio — under stress
Andina works with 4 banks. Each has its committed line, current drawdown, and covenant. Move the EBITDA stress and watch each relationship's health change. Real free cash isn't what you have — it's what you'll STILL have if things get hard.
EBITDA stress
−0%
0% = normal year. 30% = mild recession. 50% = crisis. Each bank reacts differently based on exposure and covenant.
Total committed
$155M
Total drawn
$80M
Total available
$75M
48%
Available under stress
$50M
What you can actually use if the modeled stress arrives. Banks in red freeze the line — their available does NOT count.
Total capacity
Banco Principal
Tenure: 12 yrs
Banco Secundario
Tenure: 8 yrs
Banco Estatal
Tenure: 4 yrs
Banco Inversión
Tenure: 2 yrs
What you are seeing
Three critical lessons: (1) "Committed line" does not mean "available under stress". When EBITDA drops 30%+, covenants tighten and banks freeze facilities — exactly when you need them most. (2) Diversify BY BANK: 4 relationships of $25M each is safer than 1 of $100M, because if the big bank gets difficult you lose everything. (3) The honest metric is not "total lines," it's "lines available under modeled stress." That number is typically 40-60% of the headline. The CFO who reports the headline to the board is selling air.
The critical reading of the visual: diversification is NOT just number of banks, it is DIVERSITY of profiles. 4 private banks with similar covenants (3.0-3.5x) react the same under stress. True diversification includes banks with different appetites: traditional private, investment bank (more expensive, faster), state or multilateral bank (slower, more patient), or even alternative facilities (institutional factoring, ABL revolver) — each with its own dynamic.
The second reading: "relationship tenor" matters more than theory recognizes. The bank that has known you for 12 years, that has seen a couple of cycles with you, does NOT act like the bank that has known you for 2. Under stress, relationship age translates into patience and willingness to renegotiate terms instead of cutting. That's why it pays to start banking relationships with a 10+ year vision, not transactional.
And critical: "uncommitted facilities" don't count in this analysis. If your bank can cancel the facility with 30 days notice at discretion, it is not reliable liquidity — it's marketing. REAL lines are formally committed, with commitment period (typically 2-5 years) and negotiated covenants.
The mechanics: how to build mature banking relationships
- Diversify by BANK PROFILE, not just by number. Minimum 3 banks: one main (long-term, knows the business), one secondary (competition + backup), one alternative (state/multilateral/development bank, patient in stress). If you only have "the big bank and nothing else", you're exposed to a single decision of a single committee.
- Negotiate COMMITTED LINES, not uncommitted. The difference is nominal in fees (commitment fee 0.25-0.5%/year on the undrawn portion), but structural in trust: a committed line is not unilaterally canceled. An uncommitted one is. If your bank only offers uncommitted, find another that offers committed for the same amount.
- Fight covenants before you need them. Covenants are negotiated at original contract signing, NOT during a stress event. Ask for thresholds 30-40% more generous than what your base model supports — that gives you cushion to survive 1-2 bad years without breach. Under stress, renegotiating covenants costs amendments with fee of 0.5-1% of principal + pressure on terms.
- Report proactively to the bank — the first week of fiscal close. Good numbers: you earn credibility for future renewals. Bad numbers: the bank hears from you, not from news or overdue accounts. The relationship manager is your internal ally if informed on time; your adversary if they discover the problem late.
- Keep each bank at max 35-40% of your total credit. Higher concentration makes you vulnerable to a single credit committee decision. If your main bank changes sector exposure policy or rotates the manager who knows you, you lose that whole chunk of capacity in a week.
- Cultivate the relationship BEYOND the commercial manager. Ask for annual meetings with the regional director. If your company is material, with the head of corporate banking. That relationship protects you when the commercial manager rotates — and they rotate every 2-3 years in big banks.
- Document your banking "crisis playbook". Before any stress moment, write: which banks will renew first (the most loyal)? Which will create obstacles (the newest / with tight covenants)? What alternative pre-approved facilities do you have? That document, written in calm, saves you weeks in panic.
- Committed line. The bank is contractually obligated to disburse up to amount X during period Y, subject to covenant compliance. It is reliable liquidity.
- Uncommitted line. The bank CAN disburse but is not obligated. Can cancel with short notice. NOT reliable liquidity under stress.
- Commitment fee. Annual fee (0.25-0.5%) on the UNDRAWN portion of the committed line. What you pay for guaranteed availability. Worth every cent.
- Financial covenant. Condition the company must maintain (Net debt/EBITDA under X, Interest coverage over Y). If breached, the bank can accelerate the loan, freeze the line, or charge penalty.
- Operational / information covenant. Obligation to deliver financial statements, quarterly reports, material event notifications. Easier to comply with but breach also triggers technical default.
- Cross-default. Clause stating "if I default with any other creditor, I also default with you." If one bank triggers cross-default, all your creditors enter accelerate position. The worst-case scenario.
Adversarial check
Adversarial check
1.Your main bank offers $80M uncommitted line at 25bps less than your current bank offers committed for $70M. Which do you take?
2.Why does diversifying by BANK PROFILE matter more than diversifying by NUMBER of banks?
3.Your base model shows Net debt/EBITDA at 2.3x and your main bank offers covenant at 2.8x. Do you accept?
Exit checklist
Suggested re-review: quarterly with treasury executive review, annually with renewal negotiations. An early conversation with the bank about significant events of the year (M&A, large capex, ownership change) is worth more than any rushed negotiation later.
Optional
Go deeper
Sources and books to dig into the original material