PolicyScope Data -- The Use Cases

When investment professionals realize they can measure and anticipate policy-related market volatility using PolicyScope data, they ask us: how can I use the data?


The first answer is simple: you are already incorporating public policy risk into your pricing and risk measurement. Every time portfolio managers adjust a parameter setting or assumption within pricing and risk models, they translate public policy risk into a factor used in the pre-trade investment process.


But the current process for incorporating public policy into pricing models is haphazard, episodic, implicit, and often driven by a reaction to headlines. It can be tainted by analyst bias. Finally, the reaction function feeds the misperception that public policy risk is a random variable.

PolicyScope data addresses these shortcomings by making it possible to standardize and systematically price public policy risks. In addition, our recent backtests indicate that the dataset generates powerful leading indicators of market volatility precisely because it facilitates finding developments before the broader market has had an opportunity to price it in.


Informational advantages from our patented quantification process accelerate alpha discovery and deliver to volatility traders a particularly powerful alpha finding mechanism.

In other words: multiple use cases exist even before we start talking about anticipating the actual outcome of a public policy process or decision. Measuring and pricing against volatility makes it possible to discover and capture previously hidden alpha. No crystal ball necessary.

So now, without further ado, here are……


The Top 3 Uses of PolicyScopeTM Data in Quantitative Finance


Measuring Systematic Risk (anticipate volatility): The recently released backtest results make clear that public policy risk is not a random variable. Concrete correlations have been shown in relation to four representative lexicon terms (trade war, cryptocurrency, CBDC, and LIBOR), with an average 10-22 days advance notice. Volatility traders and their risk managers can therefore use PolicyScope data to generate trading signals that capture previously invisible alpha opportunities.

Increased activity and/or increased volatility in measured public policy risks using the patented process that generates PolicyScope data can be configured by volatility traders as an opportunity to take positions at least in the S&P and the VIX at an early stage.


It is very true that markets are efficient….but they are not immediately efficient. It takes time for markets to notice and price in the implications of a particular policy activity. Our backtests indicate the time lag averages between 10-22 days.


Volatility traders to not have to know the details of the policy move. They don’t even need to know whether the activity is normatively good or bad for an economic sector. They just need to know that increased activity is underway; market reactions will follow once the information filters through to markets through the news cycle and expert analytical cycle. PolicyScope data delivers powerful informational advantages to volatility traders as a consequence because it delivers to them data-driven, fact-based foundations for pre-trade decisions that anticipate market volatility reliably.


Factor Analysis (just the data) – find the covariances: For the first time, portfolio managers and risk managers have access to a structured data set that supports systematic and detailed analysis of how public policy activity and market activity relate to each other.


Most market participants currently view public policy risks as orthogonal to market risks (i.e., statistically independent despite periodic intersections). They treat policy risks as a random variable not capable of measurement or anticipation because they predominantly experience public policy risks through the prism of volatile headlines. But they know that policy risks can and do drive asset prices.


Assume for the moment that you continue to view public policy as a random variable despite the recently released PolicyScope backtest. The structured format of the dataset (including the expressions mapped to economic sector) makes it possible to incorporate public policy risks systematically into any factor analysis framework to identify covariances among observed random variables.


In other words: PolicyScope data provides the capacity to start factoring in public policy risks explicitly into pricing models systematically, delivering superior mechanisms for identifying previously hidden covariances between portfolio exposures and public policy activity. With 1000+ lexicon terms mapped to 300+ economic sectors, the opportunities to identify covariances and correlations across a range of issues and portfolio positions related to public policy risks are significant.

The most immediate gains in this context seem likely to be captured by thematic and sectoral investors. Earlier this spring, BCMstrategy, Inc. mapped its lexicon terms to 300+ economic sectors. Some mapping examples appear below.

Thematic and sectoral analysts therefore can see daily activity level volumes and time series data expressed in terms of economic sector. Economic sector daily volumes represent the total of all public policy activity relevant to that sector across issue areas. If relatively obscure public policy move occurs that matters to the sector, the mappings surface that data as a volume measurement.


Human PolicyScope users relying on interactive dashboards additionally can drill down to read the underlying documents pertaining to those activities so they can make expert judgements regarding policy directionality.


Later this year, Bloomberg Terminal users will have access to an app that delivers interoperability and drill-down capabilities for PolicyScope data in relation to market data, portfolio positions, news sentiment analysis, and Bloomberg Intelligence analysis.


Scenario Analysis and Nowcasting (anticipate market moves and risk): PolicyScope data can also be used to augment scenario analysis. Portfolio managers and risk managers use scenario analysis to measure expected shifts in portfolio values based on observed or hypothetical changes in specific parameters and assumptions. The nowcasting variant of this activity uses flash estimates and other near-term data to generate near-term risk forecasts.


Adjusting scenario model assumptions and parameters can be a mechanical exercise. For example, monthly updated Purchasing Manager Index (PMI) or Consumer Price Index (CPI) or Gross Domestic Product (GDP) datapoints can be automatically incorporated into a spreadsheet or pricing model which automatically generates a new risk estimate based on the new data. The

For example, consider a scenario analysis for a portfolio of exposures in the technology sector. PolicyScope measurements detect a sharp increase in activity either for the sector as a whole or for a specific lexicon term particularly relevant for the sector (e.g., rare earths or data privacy or digital services tax). Increased activity signals increased risk of policy shifts, warranting a shift in model assumptions and parameters.


Scenario analysis architects can make this part of the analysis automatic by identifying in advance the baseline scenario (e.g., no change in policy). As they do with consumer prices or GDP data, they can also establish a rule that automatically registers an increase in risk of a shift from the baseline associated with public policy volume levels. Increased activity can signal increased risk because the more policymakers act and talk about making changes, the more likely it is that change will occur, on average. More sophisticated signaling related to the dynamics around public policy volumes (e.g., duration, the slope of the line, etc.) are suggested HERE.


Expert human judgement also plays a role in scenario analysis adjustments. Chief economists and strategists assess new data releases and recommend alternative parameterizations. They incorporate implicitly their assessment of public policy risks associated with a given scenario already. PolicyScope data makes it possible for the public policy portion of the analysis to become both explicit and quantitative in nature.


Consider again the example above, in which increased measured volumes of public policy activity trigger a shift in risk expectations within a scenario model. Chief economists and strategists responsible for monitoring and adjusting those models do not just view the data in a vacuum. They read a range of technical materials as well as the news cycle. They will have views on the directionality associated with any given increase in public policy activity.

Scenario analysis utility functions increase exponentially when chief economists and strategists use PolicyScope data to enhance scenario model parameterization and assumption adjustments.

Comparing daily quantified activity levels in PolicyScope datasets with the news cycle supports smarter, data-driven decisions faster about model parameterization. Evaluating time series data delivers superior pattern identification regarding the public policy reaction function.

With 1000+ lexicon terms spanning a wide range of public policy issues from monetary policy and trade policy to climate-related risk, digital currency policy, and banking regulation, the PolicyScope data set supports far more targeted data-driven scenario analysis and nowcasting analysis.

 

The full PolicyScopeTM dataset is available to institutional investors exclusively through the Bloomberg Enterprise Access Point. BCMstrategy, Inc. also delivers sector-specific datasets and related signals via API, FTP, and interactive dashboards. Coming soon: a Bloomberg Terminal App that delivers interoperability between PolicyScope data for selected issues and market/portfolio data as well as Bloomberg Intelligence.