BackStartups

Time and cash

The J-curve

Why a portfolio looks worse before it looks better, and how to tell an ordinary J-curve from a portfolio that is genuinely not working.

Plot the value of a venture portfolio against time and the line usually falls before it rises. Fees and early write-offs arrive first; the gains arrive years later. The resulting shape is the J-curve, and it is a structural feature of the asset class rather than a sign of poor selection.

For an angel the curve is shallower than for a fund, because there are no management fees dragging the early years down. It is still there, driven by the timing of failures against the timing of exits: losses resolve in years three to six and successes resolve in years seven to twelve.

The practical importance is psychological. The bottom of the curve is where angels quit, and quitting at the bottom means paying for the losses without collecting the gains.

The shape and its causes

Three things create the curve. Losses resolve faster than wins, because failing takes two to four years and succeeding takes eight to twelve. Early valuations are conservative, because marks are set by the last round rather than by any assessment of progress. And any fees or carry in a syndicated structure are charged before any gains arrive.

The result is a portfolio that shows a declining paper value through the middle years while the underlying position is improving. The companies that will produce the return are quietly compounding and being carried at their last round price, while the ones that failed have been written to zero.

The shape of an illustrative angel portfolio over twelve years
YearWhat has happenedReported value
1-2Capital deployed, some up roundsAround cost or slightly above
3-5Failures resolve, winners carried at last roundBelow cost — the bottom of the curve
6-8First exits, winners raise at higher pricesRecovering
9-12The largest positions resolveThe result, whatever it is

A constructed illustration of the pattern rather than a projection of any actual portfolio.

Telling a J-curve from a failure

Both look the same on a value chart, so the distinction has to come from the underlying companies rather than the number.

A healthy portfolio in the trough has surviving companies that are growing, raising at higher prices from new investors, and reporting regularly. The write-offs are companies that failed, not companies that went quiet. There is at least one position that could plausibly become very large.

A portfolio that is genuinely not working has surviving companies that are flat, raising only from insiders at the same or lower prices, and reporting erratically. Nothing in it has outlier potential. The distinction is usually clear once you look at the companies rather than the total.

Living with it

Do not mark your own portfolio frequently. Private company values are set by financing events that happen every eighteen to twenty-four months, so a quarterly valuation exercise generates anxiety without generating information.

Do keep deploying through the trough. Vintage diversification is one of the few genuinely free improvements available, and stopping in the middle years concentrates your capital in whatever the market was doing when you started.

And judge the portfolio on the surviving companies rather than on the number. In year four the number is a statement about which outcomes have resolved, not about what the portfolio is worth.

In short

What to take away

  • Expect the portfolio to look worst in years three to five. That is the structure, not a verdict.
  • Judge the trough by the health of the surviving companies, not by the total value.
  • Keep investing through the middle years — stopping concentrates your vintage exposure.
  • Mark rarely. Frequent valuation of illiquid positions produces anxiety, not information.

The J-curve: common questions

Why does a venture portfolio lose value before it gains?
Because failures resolve in two to four years while successes take eight to twelve, and surviving companies are carried at their last round price rather than at their progress. The losses land early and the gains land late, which produces the J shape.
How do I tell a normal J-curve from a portfolio that has failed?
Look at the surviving companies rather than the total. Healthy survivors grow, raise from new investors at higher prices and report regularly. A failing portfolio has survivors that are flat, funded only by insiders, and nothing with outlier potential.
Should I stop investing during the trough?
No — that concentrates your entire portfolio in one vintage, which is a risk you are not paid for. The middle years feel like the worst time to invest and are frequently among the better ones, because prices are lower when everyone else feels the same way.