CAPE Relative-Value Momentum Filter
Buy historically cheap sectors while excluding the weakest value trap
- Difficulty
- Advanced
- Time to result
- ~months to results
- Steps
- 6
- Confidence
- 99%
The process starts with sector-level CAPE ratios but does not compare their raw values directly because sectors naturally trade at different valuation ranges. Instead, each sector's current CAPE is compared with its own historical average, creating a normalized measure of relative cheapness. The sectors are ranked monthly, and the five cheapest become candidates. A momentum screen then removes the worst-performing candidate, leaving four sectors in the portfolio. Valuation supplies the opportunity set, while momentum helps avoid a sector that is cheap because its deterioration is accelerating. The approach treats CAPE as an indicator of prospective return rather than a precise market-timing signal and allows factor characteristics to arise from the rules rather than targeting factors directly.
Origin
Robert Shiller and Barclays developed the investable CAPE index concept, which DoubleLine analyzed and paired with actively managed fixed income exposure.
Core principles
- 01Compare each sector with its own history
- 02Avoid structural valuation bias across unlike sectors
- 03Use value to identify candidates, not market timing
- 04Use momentum to remove the weakest candidate
- 05Let factor exposure emerge from the process
How to run it
- 1
Build Sector CAPE Measures
Calculate a ten-year, inflation-adjusted earnings valuation measure for every eligible sector. Use consistent and sufficiently reliable earnings histories.
Pro tip Document accounting and sector-classification changes across the historical period.
Watch out Weak historical earnings data can make the comparison unreliable.
- 2
Normalize Each Sector
Compare every sector's current CAPE ratio with its own historical average or trading range. This prevents naturally low-multiple sectors from dominating the selection.
Pro tip Use the same normalization method across all sectors.
Watch out Comparing raw multiples across structurally different industries introduces bias.
- 3
Rank Relative Cheapness
Rank the normalized sectors monthly from cheapest to most expensive. Retain the five cheapest as the value candidate set.
Pro tip Fix the ranking date and rebalance schedule in advance.
Watch out Do not interpret the ranking as a forecast of an immediate market turn.
- 4
Apply the Momentum Filter
Measure recent performance among the five candidates and remove the worst performer. This screen is designed to reduce exposure to cheap assets still deteriorating rapidly.
Pro tip Keep the momentum definition stable to avoid discretionary overrides.
Watch out Momentum can temporarily exclude a sector shortly before it rebounds.
- 5
Allocate to the Remaining Four
Construct the equity exposure from the four surviving sectors according to the index rules. Monitor concentration and implementation costs.
Pro tip Use transparent weighting rules that can be reproduced.
Watch out Do not add discretionary sector bets while claiming to follow the systematic process.
- 6
Repeat Monthly
Refresh valuations, relative rankings, and momentum at each monthly review. Track whether turnover and realized costs remain consistent with expectations.
Pro tip Audit every inclusion and exclusion against the published rules.
Watch out Excessive methodological changes can turn the process into an overfit backtest.
In the wild
Energy appeared extremely cheap relative to its own CAPE history in 2014. As oil and energy equities declined, the momentum filter eventually removed energy from the candidate set, leaving the other four sectors rather than continuing to buy the falling sector solely because its valuation looked inexpensive.
→ Sherman said the exclusion saved approximately 600 basis points over the following couple of years compared with focusing only on the five cheapest sectors.
Common mistakes
Comparing Raw Sector Multiples
Utilities and technology have structurally different valuation ranges, so raw CAPE comparisons can create persistent sector bias.
Using CAPE as a Sell Signal
CAPE informs prospective returns but does not specify when an expensive market will fall.
Buying Every Cheap Sector
Pure valuation can keep adding to a rapidly deteriorating value trap; the momentum filter exists to remove the weakest candidate.
Is it for you?
Best for
It is best for systematic investors seeking a repeatable sector-allocation process that combines valuation and momentum.
Not ideal for
It is not ideal for markets lacking credible, consistent ten-year earnings histories at the sector level.
From the transcript
“So to normalize this, or to try to standardize it, what you do is you compare each sector's cap ratio to its own historical average.”
“So you rank these each month, and you're gonna pick the bottom five.”
“So the five sectors, whicheveryone is the worst performer, you throw it away.”
From the episode
Jeffrey Sherman on Trading in Academics for Fund Management
Jeffrey Sherman