Deciding between holding cash and repaying debt
Compare the after-tax cost of debt with what your liquidity actually earns, then test the cash against the seasonal peak. Official sources give the obligations, not the numbers.

Most companies with spare cash and a loan on the books ask the same question at the same meeting: should we pay the debt down? The instinct is usually sound, because debt that costs more than the cash earns is a drag on the balance sheet. The trouble is that the instinct ignores the calendar. A business that repays its loan in March and then cannot meet its tax payment in June has not saved money; it has borrowed again, probably at a worse moment. This framework sets out four tests to run before the paydown decision goes to the board. It relies on institutional guidance and on what the sources actually say. Where the sources are silent, the judgement is ours, and we say so.
Chapter one: Price the debt after tax
Start with the true cost of the loan, not the rate printed on the term sheet. The all-in figure includes fees, any commitment charges, and the interest rate as the agreement defines it. The loan agreement is the only reliable source for those terms, and the definitions section usually matters more than the headline. Once you have that number, the after-tax version is the one to compare with your alternatives. The adjustment depends on how your jurisdiction treats interest for your business, and that is a question for an accountant or tax adviser, not for a treasury spreadsheet. We do not supply a tax rate or a method here, because the sources reviewed for this piece do not contain one.
A common error is to compare the after-tax cost of debt with the pre-tax yield on cash, as though the two were measured the same way. They are not. Interest paid is a cost you can often deduct; interest earned is income you may be taxed on. Put both on the same basis before you compare them. In our experience, the cleaner habit is to write the net figure for each option on one page, with the assumptions named, so that the board can see which number moved when someone changed a line.
Chapter two: Price the cash you would hold
The return on holding liquidity is the number treasurers most often overstate. The advertised deposit rate is a starting point, not an answer. Deduct any account fees, minimum-balance conditions and the cost of moving money when you need it. Then ask whether the rate will still apply in the month you need the cash, because a rate that falls or a balance that drops below a threshold changes the answer. The sources we reviewed give no interest rates and no yields on cash, so there is no external benchmark in this article to check your figure against. The benchmark has to come from your own bank and your own agreements.
Central bank decisions affect this comparison, but the sources here do not let us say how. The Federal Reserve's FOMC statement dated 28 October 2026 is listed in our fact sheet, but the extract we were given contains no rate or policy content, so we cannot report what it did. The Fed has also announced a revised tentative 2027 meeting schedule, on 12 November 2026. A sensible habit is to add those dates to the treasury calendar so that the cash forecast is reviewed around them. Beyond that, the rate outlook belongs in the model as an explicit assumption, labelled with its source and date, and not as a line of unexplained optimism.
Chapter three: Size the seasonal buffer
This is the test that most paydown decisions skip. A buffer is the cash you hold so that known outflows can be paid on time, whatever happens to receipts. The size of the buffer should be set by the worst dated month in the next two to three quarters, not by the average balance across the year. Take the calendar of outflows and find the peak. For a business that pays quarterly estimated income taxes, that peak may fall on a quarterly due date. For an employer, withheld income tax and Social Security and Medicare taxes must be paid over as the payroll runs, so the payroll cycle is itself a fixed outflow. Add seasonal inventory or contract payments where they apply. Then ask a plain question: if receipts stopped for one month, could you still pay everything that falls due in the peak month? If the answer is no, the buffer is too small, and paying down debt would make the shortfall worse.
The sources give us the obligations but not the sizing rule. The IRS states that estimated taxes are paid quarterly and based on the income a business expects to earn in the year, and that federal income tax is paid as income is earned. Those facts tell you when money leaves the account. They do not tell you how many weeks of cover to hold, and we would be inventing a number if we suggested one. The Bank for International Settlements publishes statistics compiled with central banks to inform analysis of global liquidity, and its Basel III summary describes the post-crisis reforms. The extract we reviewed does not say what liquidity requirements apply to a given firm, so any regulatory buffer is a question for your bank and your regulator. The internal buffer test above is our judgement, and the board should own its parameters.
Chapter four: Map the obligations you cannot defer
Some payments can be moved with a phone call; others cannot. Before you decide on a paydown, sort every outflow into two groups: those you can defer and those you cannot. Employment taxes and estimated income taxes belong in the second group. The IRS provides an online account where taxpayers can view their balance and make payments, and it offers payment plans and a payment-option finder for tax debt. Those are useful when a payment is missed, but they are a remedy after the fact, and the sources do not give the interest or penalty rates that apply to them. Read the current IRS terms before you rely on a plan as a backstop, and treat any unpaid tax as a debt with its own cost.
The second half of this pillar is the exit. Ask whether the cash you are about to commit to repaying can be redrawn, and on what terms. A prepayment on a term loan may be irreversible, while a revolving facility may be cheaper to repay and to re-draw later. Check the loan agreement for prepayment conditions and for any requirement to keep a minimum balance. If a repayment cannot be reversed inside the peak window, mark it as a decision for the board rather than a treasury action.
Before the paydown goes to the board
Answer each question in writing. Any no means the paydown waits.
- Is the after-tax cost of the debt calculated from the all-in rate in the loan agreement, a
- Is the return on cash stated net of fees, minimum-balance conditions and access costs, wit
- Has the worst dated outflow month in the next two to three quarters been identified, inclu
- After the proposed paydown, can the business still pay that peak month if receipts stop fo
- Can the amount repaid be redrawn inside the peak window, and do the loan terms say so?
- Have the central bank meeting dates in the forecast period been added to the treasury cale
- Has the board minuted the buffer test and its assumptions, so that the next review can che
Outgrace checklist
Where it breaks
The framework fails in three predictable places. The first is the forecast. A buffer sized to a peak that has been underestimated is no buffer at all, and the usual cause is a receipts forecast that assumes the peak will arrive when last year's did. Rebuild the outflow calendar from the tax and payroll obligations themselves, not from the previous year's bank statements. The second is the assumed return on cash. Treasurers who take the headline deposit rate into the model will overstate the case for holding cash and may then be surprised when the balance does not earn what they expected. The third is the investment side. Companies that hold long-term corporate bonds as part of their liquidity are exposed to a different risk. The European Central Bank says that US hyperscalers have rapidly raised their share of long-term euro corporate bonds, that growing hyperscaler presence makes investors more exposed to a few issuers, and that this exposure could amplify portfolio losses if AI-related revenues disappoint. The ECB describes that analysis as early stage. A liquidity reserve held in bonds is not the same as a liquidity reserve held in cash, and the buffer test should say which one is being counted.
There is a further limit. The framework tells you how to decide, not what the answer will be. Some businesses will find that the after-tax cost of their debt is higher than the net return on cash and that the buffer is comfortably covered; the paydown is then straightforward. Others will find that the buffer is thin and that the debt is cheap. In that case, the honest answer is to hold the cash and to ask the lender whether the repayment can be staged. We have not invented a rule that favours one answer over the other, because the sources do not support one. The outcome depends on your numbers, and the test is only as good as the inputs that go into it.
On Monday, start with the outflow calendar, not the loan. Write down every payment that will leave the account in the next two to three quarters, mark the ones you cannot defer, and find the worst month. Then pull the loan agreement and find the all-in rate and the prepayment terms. Ask your bank for the net return on the cash you hold, in writing. Send the three documents to your finance lead and your accountant, and ask the accountant to confirm the tax treatment of the interest. Bring the one-page comparison to the next board meeting with the buffer test at the top. If the peak month is covered after the proposed paydown, the decision can go ahead. If it is not, the paydown waits until it is.