A US equity feeder fund lagged its index by 2.3% in a year: expense ratio 0.6%, the overseas fund's own cost 0.2%, 5% held in cash while the index rose 20%, and currency moves timed differently from the NAV cut-off. Decompose the gap, and explain what happens when the industry's overseas investment limit is reached.

The 2.3 points split into 0.6 of expense ratio, 0.2 of overseas fund cost, 0.70 of cash drag and about 0.80 of currency and timing differences. The fees recur every year; the cash drag depends on the market and was largest because the index rose 20%; the timing residual is noise that can reverse. If the overseas limit is reached, the fund stops taking fresh money and SIPs, existing investors stay invested, and listed overseas ETFs can trade at a premium to NAV.

Four of them, and they behave differently. Buying a product through an importer and then a local shop means paying two margins, waiting while some stock sits in the back room, and converting the price at whatever rate the shop used that morning. A feeder fundA fund of funds that invests almost all its assets in a single other fund, here an overseas index fund, so its investors carry the costs of both. carries two layers of fees, a cash buffer that earns a different return from the index, and an exchange-rate timing difference. The fees are 0.6% for the feeder and 0.2% inside the overseas fund's NAV, a combined 0.8% that is lost every year whatever the market does.

Price it as a missed return. The 5% in cash earned about 6% while the index made 20%, so it cost 5% x (20% - 6%) = 0.70 points, more than the feeder's own expense ratio. That is the claim distributors miss: in a strong year, the cash costs more than the fee. But it is not a fixed cost. The drag is zero when the index returns the same 6% as cash, and in a falling year it turns into a cushion. In a year the index lost 20%, the same cash would have added about 1.3 points. The cash is a small, permanent bet against the market.

After fees and cash, 0.80 points remain. The benchmark converts the US index into rupees at one exchange rate, while the fund strikes its NAV using the overseas fund's price and a rate taken at a different time. On any day the rupee moves between those two moments, the fund and the benchmark disagree. Over a year these differences mostly net out, so a residual of 0.80 is more likely noise than a leak, but a residual of the same sign every year would be a process problem worth raising. The right check is the residual's history: three years of differences that wander around zero are noise; three years that are all negative are not.

The regulator caps how much the whole mutual fund industry, and each fund house, can invest abroad. When the cap is reached, a feeder cannot buy more of the overseas fund, so the fund house stops accepting fresh lump sums, new SIP registrations and often existing SIP instalments, and switches into the scheme. Existing investors stay invested and can redeem as usual, but anyone who leaves cannot easily come back, and SIP plans break. For overseas ETFs listed in India, creation of new units stops, so buying pressure on the exchange pushes the market price above the ETF's NAV; buyers then pay a premium that disappears when the limit is relaxed. A fund that keeps accepting money it cannot invest would also raise its cash share, adding to the drag. Confirm the current limits, per fund house and for the industry, before quoting any number.

Close with a view for the distributors. The feeder's structural lag is about 0.8 points a year in fees plus the cost of its cash, which in strong years can add as much again. The fund house can narrow it by holding less cash, using a lower-cost overseas share class, and checking that the timing residual stays around zero. The limit to say: one year is a single draw of the cash and timing effects, so the decomposition should be repeated over several years before anyone calls the fund badly run.