Factor winters: the premium is pay for the drought

A friend forwarded me a value tilt fund last week. Few years in, it had gone nowhere, and he wanted to know if the whole idea was broken.

I pulled the long history before answering.

Value, measured as a factor, went flat to negative from 2007 all the way to 2020. Thirteen years. At the bottom, the drawdown was somewhere around 55 percent. That is the deepest and longest drought the factor has shown since 1963. Every smart beta brochure still quotes the long run premium, around three to four percent a year. Almost none of them quote the thirteen year wait that came attached to it.

Here is the part that catches people. A 55 percent hole needs a 122 percent gain just to climb back to flat. Value did recover, sharply, across 2021 and 2022. But whoever gave up at the bottom locked in the loss and handed the entire rebound to whoever was still holding.

What took me a while to really get is that the drought is not a flaw in the factor. It is the reason the factor pays at all. If a premium stopped hurting in the dry years, everyone would crowd in and arbitrage it flat. The discomfort is the toll. You are being paid to hold the thing other people cannot stomach holding.

A winter is different from a crash too. A crash is sharp and over fast. A winter just grinds, quarter after quarter of nothing, and that slow ache is what shakes investors out at exactly the wrong moment.

So before you buy any factor, ask one question. Can I sit through its worst winter without selling. Size it so a flat decade annoys you instead of breaking your plan.

A factor does not pay you for being right. It pays you for staying when the chart says you were wrong.

The arithmetic of a hole

The 122 percent in the opening is not a typo. A fall and the gain needed to undo it are not the same size, because the gain is measured from the lower base. Lose 55 percent and Rs 1,00,000 becomes Rs 45,000. To get back to Rs 1,00,000 you need Rs 55,000 more, and Rs 55,000 on a base of Rs 45,000 is 122 percent.

The time it takes depends on the pace of the climb. At 10 percent a year, a steady rate chosen for the arithmetic, that climb takes about 8.4 years. At 15 percent, about 5.7.

Try 30.7, the deepest fall of an Indian factor index in the last two years, which comes up below. It needs 44.3 percent to recover. Smaller hole, same shape of problem.

Indian factors are in a winter right now

This is not only a story about US value between 2007 and 2020. Look at NSE's own factor indices over the two years to 9 Oct 2026, price only.

The Nifty200 Momentum 30 fell 30.7 percent from its high to its low inside those two years, and on 9 Oct 2026 it was still 21.5 percent below where it started. Quality is the slower case, the true winter. On NSE's factsheets for 30 Sep 2026 the NIFTY200 Quality 30 earned 6.40 percent a year on total return over five years, against 8.30 for its parent Nifty 200. Since April 2005, back calculated years included, the same index shows 16.64 percent against 13.46. A long run lead, and a five year stretch of trailing inside it.

That is what a winter looks like from the inside. No single terrible day. Just a factor that keeps finishing behind the plain index, year after year, while the long run number on the factsheet still looks fine.

Crash or winter: why the second one does more damage

The opening draws the line between the two. Here it is side by side.

A crash tests nerve for a few weeks. A winter tests belief for years, and belief is what gives way first. The investor who sells in a winter rarely sells at a dramatic low. They sell after the third flat year, when the reason for holding stops feeling true.

The free module on factor timing and cycles goes through why these cycles run long and irregular, and why exiting after two bad years has historically meant missing the payoff. For the value side, which value you hold changes how deep the winter gets, and the value trap post covers the stocks that are cheap for a reason.