Fundamental Selection with Quantitative Risk Control
Select companies bottom-up, then expose and control hidden portfolio risks
- Difficulty
- Advanced
- Time to result
- ~months to results
- Steps
- 5
- Confidence
- 98%
The method combines two complementary lenses. Fundamental analysts investigate businesses, management, governance, financial strength, valuation, and company-specific risks to select securities. Quantitative colleagues then disaggregate the resulting portfolio into systematic exposures such as cyclicality, volatility, value, and size. Those findings reveal concentrations that bottom-up analysts may not see when evaluating companies individually. The portfolio can then be adjusted so that returns depend more on deliberate security selection and less on accidental factor bets. The information also flows in reverse: direct company research and governance checks can improve broad quantitative portfolios, especially in emerging markets where accounting quality and management behavior may be difficult to infer from raw data alone.
Origin
Ketterer developed the approach after observing in the late 1990s that competing fundamental managers often did not understand their systematic portfolio risks. Causeway Capital was subsequently built as a convergence of fundamental investing and quantitative research.
Core principles
- 01Fundamental research identifies company-specific opportunity.
- 02Quantitative analysis reveals systematic risks hidden across holdings.
- 03Diversification should result from measured risks, not arbitrary quotas.
- 04Human company research can improve statistically constructed portfolios.
How to run it
- 1
Build the fundamental case
Analyze each company’s assets, management, governance, financial strength, earnings potential, and valuation. Identify the company-specific thesis and failure conditions.
Pro tip Use direct management meetings to test governance and capital-allocation assumptions.
Watch out Do not mistake management confidence for independent evidence.
- 2
Measure systematic exposures
Use quantitative analysis to decompose the proposed portfolio into factors such as cyclicality, volatility, value, and size. Distinguish these shared exposures from company-specific risk.
Pro tip Evaluate the whole portfolio rather than reviewing factor risk one stock at a time.
Watch out A collection of attractive companies can still create a concentrated macro bet.
- 3
Control unintended risks
Resize or replace holdings that create excessive factor concentrations while preserving the strongest fundamental ideas. Seek diversification as an outcome of explicit risk control.
Pro tip Focus on unintended exposure rather than mechanically neutralizing every factor.
Watch out Over-optimization can eliminate the active views that justify the portfolio.
- 4
Feed research back into the model
Use information from company meetings and governance analysis to challenge or refine quantitative signals. This is especially important where the statistical portfolio contains many less-followed companies.
Pro tip Flag companies whose governance quality makes their reported data less dependable.
Watch out Do not assume clean numerical inputs represent clean underlying businesses.
- 5
Repeat as conditions change
Update company estimates, prices, and factor exposures as new information arrives. Rebalance when either the fundamental thesis or the portfolio-level risk profile changes.
Pro tip Review exposures after large market moves, when correlations and rankings may shift quickly.
Watch out A risk analysis becomes stale as soon as prices and business conditions materially change.
In the wild
An international value manager finds attractive banks, automakers, and chemical companies. Fundamental analysis supports each holding, but factor decomposition shows that the combined portfolio carries excessive cyclical and European macro risk. The manager retains the highest-conviction names, reduces overlapping exposures, and adds unrelated opportunities with comparable expected returns.
→ The portfolio preserves its value thesis while reducing unintended sensitivity to one economic scenario.
A quantitative screen identifies a broad set of statistically cheap emerging-market stocks. Analysts meet selected management teams and discover that several firms have weak governance or questionable capital allocation. Those findings are incorporated into the portfolio instead of relying exclusively on the screen.
→ The statistical strategy avoids companies whose apparent value is undermined by governance risk.
Common mistakes
Analyzing holdings in isolation
A stock can look attractive individually while duplicating risks already present throughout the portfolio. Evaluate aggregate factor exposure before sizing it.
Treating the model as an oracle
Quantitative outputs depend on data quality and assumptions. Fundamental evidence should challenge signals that do not fit the underlying business.
Neutralizing every active exposure
Risk control should remove accidental bets, not erase all differentiated judgment. A perfectly neutral portfolio may also lack a credible source of excess return.
Is it for you?
Best for
It is best for active investors building diversified portfolios from bottom-up security research.
Not ideal for
It is not ideal for investors without reliable data, fundamental research capacity, or tools for factor-risk analysis.
From the transcript
“our bottom up process is very much fundamental stock selection combined with quant risk control”
“our quant colleagues showed us right from the start they could disaggregate the risk.”
“fundamentally, we're actually meeting with the companies that are in that portfolio, not all, but many, making eye contact with management, probing them on their…”
From the episode
Sarah Ketterer Discusses Investment Management
Sarah Ketterer