Overnight, liquid or ultra-short: where near-term cash actually belongs

Three debt categories sit at the short end of the curve, they look almost identical on a returns table, and investors routinely pick between them by which one showed the biggest number last year. That is the wrong variable. The right one is a single question: how many days until you need this money?

The three, ranked by how far out they lend

SEBI defines each by the maturity of what it may hold, and that definition is the whole difference.

Overnight funds hold paper maturing in one day. Interest-rate risk is effectively nil — a one-day instrument cannot be repriced against you — and credit risk is minimal because the money is back tomorrow. They yield the least, and they are the calmest thing in the debt universe.

Liquid funds hold paper maturing within 91 days. Slightly more yield, slightly more sensitivity, and a residual credit question because 91-day paper can be issued by borrowers that overnight lending never touches. This is the workhorse category for a cash reserve.

Ultra-short duration funds run a Macaulay duration of roughly three to six months, so they may hold instruments maturing well beyond that. More yield again, and now a real — if small — sensitivity to rate moves and to credit events. The mechanism for both is in duration risk and credit risk are not the same thing.

The ordering is monotonic and unsurprising: the further out a fund lends, the more it earns and the more can go wrong. There is no free yield here, only progressively longer commitments.

Matching the fund to the horizon

The rule that resolves almost every case:

• Money you may need tomorrow — overnight, or the sweep facility in your bank account. The emergency reserve is the canonical case, and the fractional extra yield of a liquid fund is not worth thinking about for it.
• Money you will need in weeks to about three months — liquid. Watch the graded exit load liquid funds carry for the first seven days; it exists to discourage same-week churn and it will find you if you ignore it.
• Money you will need in roughly six months to a year — ultra-short. This is also the sensible parking spot for the source pot of an STP.

Two things that are not the deciding variable:

Last year's return. The category that led last year was usually the one taking the most duration risk into a falling-rate year. That is a statement about rates, not about the fund — and it is the classic recency bias trap applied to debt.

The tax treatment, because it is identical across all three: gains are taxed at your slab rate, with no long-term concept, whatever the holding period. That uniformity is exactly why an arbitrage fund becomes interesting for a 30%-slab investor once the horizon stretches past a few months.

The one thing to actually inspect

Since all three lend short and the tax is the same, the differences that survive are cost and credit. Check the expense ratio — on a category yielding a handful of percent, a high TER is a large share of the return — and glance at the portfolio's credit profile rather than assuming "debt" means "safe". A fund reaching for yield at the short end is reaching through credit, and what happens when a debt fund freezes is the case study for why that matters more than the extra few basis points.

⚠️ SEBI's category definitions, exit-load rules and the taxation of debt funds have all changed and can change again. Verify current rules before relying on anything here.

Key takeaway

Pick by horizon, not by last year's return. Overnight for money you may need tomorrow, liquid for weeks to three months, ultra-short for six months to a year — and accept that the extra yield further out is paid for with duration and credit risk, not conjured. All three are taxed at slab regardless of holding period, so the only differences left worth inspecting are the expense ratio and the credit quality of what the fund actually holds.