How GARP fails: the noisy denominator, the cyclical trap, and multiples that stay compressed
Every failure of this system is a failure of the number in the denominator. The arithmetic never breaks; the growth rate it divides by does, in three distinct ways.
Key takeaway
- One exceptional quarter makes PEG look wonderful right before the comparison base resets and the growth rate collapses
- Cyclicals pass the screen at the top of their cycle — peak earnings compress the P/E and inflate the growth rate simultaneously
- The re-rating the method waits for can simply not come: in growth-hostile regimes, cheap growth stays cheap for years
The denominator is a forecast wearing a historical costume
The P/E in the numerator is a fact — today's price over reported earnings. The growth rate in the denominator is presented the same way and is not the same kind of thing: buying a PEG of 0.7 is only profitable if the growth persists, so the historical rate is standing in for a forecast. Every rule in the system polices the numerator; nothing mechanical can police the denominator.
The sharpest version of the failure: a company posts one exceptional year — an acquisition closes, a competitor stumbles, a post-shortage restock. Year-over-year growth prints 40%, the PEG prints 0.5, the screen approves. Four quarters later the comparison base includes the exceptional year, the growth rate reverts to 8%, and the same stock at the same price now has a PEG above 2.
Nothing in that sequence involved the business deteriorating. The stock did not change; the denominator did. The defence is not a better formula — it is asking, for every screened candidate, what specifically produced the growth and whether that cause recurs. This is the two hours per quarter, and it is not optional.
Cyclicals look cheapest on PEG at the exact top of their cycle
At the peak of a cycle, a steelmaker or a chipmaker or a homebuilder has earnings at record highs and rising fast. Record earnings make the P/E low; the recovery from the last trough makes the growth rate high. Low multiple over high growth: the PEG can print 0.3, the most emphatic buy signal the ratio ever produces — issued at the precise moment the next earnings move is down.
Lynch's own writing is blunt about this: cyclicals are the category where the P/E playbook runs backwards, and the time to buy them is when the multiple looks terrifying, not when it looks cheap. The screen cannot tell a fast grower from a peak cyclical — both present as high growth on a modest multiple. The reader has to make that call, and the one question that makes it is: does this company's demand depend on a cycle, or on something it is doing?
The re-rating can simply not come
The method's payoff has two parts: the earnings compound, and the multiple expands toward the growth rate. The first part the company controls. The second is a claim about other investors, and in some regimes they decline to cooperate for years — rising rates compress all growth multiples, a sector falls out of favour, small and mid caps trade at a persistent discount to large.
In that regime the system does not lose money so much as it underdelivers for a long time: the earnings growth accrues while the multiple sits still, and the return is the growth alone, without the re-rating kicker the entry price implied. Positions look becalmed. The temptation is to rotate into whatever is moving — which converts a patient method into a chase, at exactly the moment its holdings are getting cheaper on their own metric.
Common questions
- Is a screen full of matches a good sign?
- Usually the opposite. The count of PEG-under-1 stocks swells in two situations: broad market falls, when it is genuine opportunity, and late in earnings booms, when it is a warehouse of peak cyclicals. Before acting on an unusually long list, check which kind of year it is — the composition of the list tells you.
- What is the single most common way people lose money with PEG?
- Buying the lowest PEG on the list. The extreme low ratio is almost always one of the two traps — a peak cyclical or a one-off growth print — because a genuinely durable 25% grower rarely gets priced at half its growth rate in a liquid market. The middle of the list, a PEG of 0.7 to 0.9 on a boring business, is where the method's actual wins live.
These are documented methods described for study. Nothing here is investment advice, a recommendation, or a claim about future returns — every system on this page has losing periods, and the pages say where.
Reading about a system is not having one.
Plutux is where you write your own rules down, test them against real data, and keep the record your memory would otherwise rewrite. Join the waitlist for early access.