Home » Research Blog » How macro-quantamental trading signals will transform asset management

How macro-quantamental trading signals will transform asset management

Macro-quantamental indicators and trading signals are transformative technologies for asset management. That is because they allow plugging point-in-time fundamental economic information into systematic trading, backtesting, and statistical learning pipelines and remove an important barrier to information efficiency. While the predictive power of macro information for asset returns has been evident for decades, its use in systematic trading and research has remained rare. This disconnect reflects the historical difficulties of replicating past data and analyses point-in-time and the need for expert curation of data updates going forward.

Macro-quantamental signals allow systematic trading to become more informed and make prices more “anchored” in economic reality. Value generation is not merely a zero-sum game but also a profit share from a more efficient financial system. The principles of quantamental success are the faster pricing of macroeconomic developments, the correction of implausible risk premia, the adjustment of evident price-value gaps, and an improved pricing of market “setback risks”. Empirical evidence has shown macro-quantamental signals succeeding in many areas, including market timing, enhancement of trend following, improvement of risk premium strategies, equity allocation, and higher-frequency information change-based strategies. Macro quantamental signals also integrate neatly with statistical learning.

What are macro-quantamental indicators and signals?

Generally, a quantamental indicator is a metric that combines quantitative and fundamental analysis to support investment decisions, such as equity valuation ratios or real interest rates. The metric must be a point-in-time information state, i.e., assign values to the date they became public knowledge, not when they occurred. More specifically, a macro-quantamental indicator is a time series of relevant macroeconomic information states designed for the development and backtesting of financial market trading strategies. Finally, a macro-quantamental trading signal is a time series of indications for market positions based on a model that involves quantamental indicators.

It is important not to confuse macro-quantamental indicators with standard macroeconomic indicators and standard macro trading signals:

  • Macro-quantamental indicators are always public information states of the latest instance of an economic state, such as GDP growth, earnings ratios, or real interest rates, as observed in real time. This means that quantamental data are based on data vintages, i.e., time sequences of complete data histories. Vintages come about through extensions of data series, revisions, methodological changes, and re-estimates of underlying models. Quantamental indicators are principally designed to replicate what the market knew at any point in time in the past and to prevent any look-ahead bias in developing and evaluating trading strategies.
  • Macro-quantamental signals are also quite different from standard systematic trading signals, which are mainly based on market prices and flows. Macro-quantamental signals directly inform on the activity, balance sheets, and sentiment of various parts of an economy. Some market data signals are also called “macro” but provide economic information only indirectly and are influenced by many other factors.

Why are macro-quantamental indicators new but not alternative data?

Macroeconomic information has been used intensely in discretionary trading for decades. However, point-in-time macro-quantamental information for systematic trading and backtesting has been rare because it has been tedious and expensive to produce at the level of an asset manager. Standard economic databases are unsuitable because they overwrite data after revisions, omit publication and revision time stamps, and have, over time, modified conventions, adjustment factors, and underlying models with hindsight.

A proper macro-quantamental indicator must, for any point in time in the past, display the value of an up-to-date economic report on that date. For example, the value of a short-term seasonally adjusted production trend on the 5th of October 2004 must be the per cent growth rate for the latest available period based on the full-time series that was published on that date based on the seasonal adjustment factor that would have been used on that day. The dates at which the growth rates are timestamped are real-time dates. The periods to which the reported information refers are observation periods.

This real-time data principle implies that even the simplest quantamental indicator is principally based on a two-dimensional data set.

  • The first dimension is the timeline of real-time dates. It marks the progression of the market’s information state. A change in values on this timeline marks updates of public information.
  • The second dimension is the timeline of observation dates. Each real-time date records the full history of an economic indicator, which is the basis for calculating the quantamental indicator, i.e., a vintage. A change in values along this timeline marks a change in an economic state according to the knowledge of the real-time date.

Hence, the underlying data set of a simple quantamental indicator (that uses only one economic time series) is a sequence of vintages. It is like an expanding data frame with a row index of progressing real-time dates and a column index of expanding observation periods. If the quantamental indicator requires more than one economic time series for calculation, such as the external trade balance-to-GDP ratio, the underlying data structure becomes 3-dimensional, with the third dimension being the index of indicators.

A quantamental database requires a large vintage data warehouse, which is hundreds of times the size of a comparable regular economic database. That warehouse must ingest information from a wide range of statistical agencies, research institutes, and data aggregators. Official vintage archives are often poorly maintained for lack of widespread interest and require quality control and “repairs”. Moreover, not all vintage information is available in archives. The progression of model-driven quantamental indicators such as nowcasts must be estimated through a sequential evolution of models and their parameters.

Due to these challenges, it was not before 2023 that the first industry-wide source of macro quantamental information for institutional investors was introduced: the J.P. Morgan Macrosynergy Quantamental System (“JPMaQS”). It is the first service that allows the systematic evaluation and trading of macro-quantamental indicators, making macro-quantamental trading strategies feasible and affordable for a much larger range of institutions.

Why are macro-quantamental indicators a game changer?

The general availability of macro-quantamental indicators means that macroeconomic information can be used in systems similar to price data, whether it is for purely algorithmic strategies or for systematically supported discretionary trading. This broadens the scope of evidently relevant features that can be used in systematic trading, giving it a new and potentially critical edge over traditional discretionary trading. While systematic signals are generally simple compared to the reasoning of discretionary managers, they can incorporate much more information than a human can. And the field of macroeconomic information is vast. At a global scale, not even specialized economists can keep up with relevant developments in all countries with sizeable financial markets. Only quantamental signals can do that.

At present, systematic trading is dominated by price and financial flow data, which are easier to understand backtest. However, macroeconomic information seems similar in importance to prices and flows.

  • Academic theory has long established that all financial contract prices are part of a broader macroeconomic equilibrium. For example, fixed-income prices reflect the demand and supply of credit and balance savings and investment at the macroeconomic level. Similarly, exchange rates balance the demand for goods and assets across currency areas. This means that if macroeconomic conditions outside the financial markets change, so must prices in the financial market.
  • Financial markets cannot possibly be fully macro information efficient. Information goes to waste due to research costs, trading restrictions, and external effects (view post here). The cost-benefit trade-off manifests as rational inattention. The theory of rational inattention argues that market participants cannot continuously process and act upon all information, but they can set priorities and choose the mistakes they are willing to accept (view post here). Moreover, research alone does not produce efficient markets. Financial markets research translates into price information only if and when it is acted upon.

The evident relevance and rational neglect of macroeconomic information explain why specialized discretionary macro trading has been so profitable over the past decades. This also suggests that macro-quantamental signals will make systematic trading more informed and more “grounded” in the economic developments outside financial markets. Hence, value generation is not merely a zero-sum game, where the better-informed trader outsmarts the less informed market participants but a profit share from more efficient financial systems and economies. Macro-quantamental strategies contribute to a core social function of the asset management industry, i.e., the provision of informative prices for the efficient allocation of scarce economic resources.

How do macro-quantamental indicators create investment value

Value generation through point-in-time macro information is not a great mystery and can broadly be classified into four principles:

  • Following macroeconomic trends: Macroeconomic trends predict asset returns for two principal reasons. They affect investors’ attitudes toward risk and influence the expected risk-neutral payoff of a financial contract. For example, an increase in inflation typically raises fear for monetary stability and increases expected future interest rates simultaneously. With rational inattention, point-in-time information on economic trends, which can encompass a wide range of indicators and countries, is never fully priced. The relevance and predictive power of point-in-time macro trends have been demonstrated in applied research for all major asset classes: fixed income, foreign exchange, equities, commodities, credit derivatives, and cross-asset return correlation (view summary here).
  • Detecting implicit subsidies: Implicit subsidies in financial markets are premia paid through transactions that have motives other than conventional risk-return optimization. They manifest as expected returns over and above the risk-free rate and conventional risk premia (view summary here). Non-profit transaction motives include monetary policy, public interventions to safeguard financial stability, convenience yields of certain assets that accrue to banks and corporates, non-standard risk aversion, and behavioural biases, such as salience bias and loss aversion. Implicit subsidies are a bit like fees for services, except that they are opaque rather than openly declared. Macro-quantamental indicators can help detect implicit subsidies by tracking plausible motives, such as deflation risk, or tracking valuations and expected returns, such as risk-adjusted real carry on currencies. Detecting implicit subsidies is information-intensive but often creates stable risk-adjusted value.
  • Detecting price distortions: Price distortions are conspicuous price-value gaps. They arise from inefficient flows and prevail as long as a sizable share of market participants is either unwilling or unable to respond to obvious dislocations. (view summary here) There are many causes of such inefficiencies, including risk management rules, liquidity disruptions, mechanical rebalancing rules and government interventions. Macro-quantamental indicators help detect distortions through fundamental valuation metrics and measures of market disruptions. For example, terms-of-trade and purchasing power parity estimates assist in tracking currency overvaluation, while bank lending surveys and central bank liquidity measures can indicate financial market stress.
  • Tracking endogenous market risk: Endogenous risk refers to uncertainty resulting from financial market positioning, as opposed to uncertainty about traded assets’ fundamental value. Endogenous market risk often manifests as feedback loops after some exogenous shock hits the market. An important type is setback risk, which refers to the asymmetry of the upside and downside potential of a trade that arises from market positioning (view summary here). Setback risk is a proclivity to incur outsized mark-to-market losses even if the fundamental value proposition of the trade remains perfectly valid. This makes it the natural counterweight to popular positioning. A useful two-factor model for detecting setback risk can be based on market positioning and exit risk. The highest setback risk is characterized by crowded positions that face an incoming shock that most investment managers have not considered. Macro-quantamental indicators help track economy-wide positioning risk, for example, by monitoring capital flows, current account deficits, and governing borrowing requirements.

Proofs of concept

There is a growing body of evidence of the relevance of macro-quantamental trading signals across asset classes, trading frequencies, and signal-generation principals. The sections below summarize this evidence to the extent that it has been documented in the Quantamental Academy.:

Market direction and timing

Simple macro-quantamental signals have displayed significant predictive power of directional returns across and between asset classes. Importantly, those signals need not be optimized; they merely follow standard economic theory and common sense.

For example, a single balanced “cyclical strength score” based on point-in-time quantamental indicators of excess GDP growth, labour market tightening, and excess inflation has displayed significant predictive power for equity, FX, and fixed income returns, as well as relative asset class positions. The direction of relationships has always been in accordance with the theoretical priors of macroeconomics (view post here). Similarly, relative central bank intervention liquidity trends, which inform on the market support of monetary policy, have been predictors of the future relative performance of assets across different currency areas (view post here). Finally, bank lending surveys help predict the relative performance of equity versus duration positions, with signs of stronger credit favouring business growth and expanding leverage. Conversely, signs of tightening credit supply bode for a weaker economy and more accommodative monetary policy, benefiting duration versus equity positions (view post here).

Enhancements of standard trend following

Classic trend following is based on market prices or returns. Market trends generate value in the presence of behavioural biases and rational herding. By contrast, macro trends track relevant states of the economy based on fundamental data. Notably, macro and market trends have complementary strengths. While market trends are timelier, macro trends are more specific in information content (view post here). Jointly, they can produce more value than individually.

Indeed, market price trends often foster economic trends that eventually oppose them. Theory and empirical evidence support this phenomenon for equity markets and suggest that macro headwind (or tailwind) indicators are powerful modifiers of trend-following strategies. For example, if one modifies standard price trend signals of equity index futures in developed markets by macro-quantamental indicators of labour markets and inflation, the resulting modified signal displays greater predictive power and material increases risk-adjusted returns (view post here).

A similar effect can be found in the foreign exchange space. A currency’s positive FX forward return trend is less likely to be sustained if concurrent economic data signal a deterioration in the local economy’s competitiveness. Empirical evidence shows that standard global FX trend following would have benefited significantly merely from adjusting for changes in external balances (view post here).

Enhancement of carry strategies

Carry can be defined as a return for unchanged market prices and is easy to calculate in real-time across assets. It is a legitimate basis for tracking risk premia and implicit subsidies. However, conceptually, carry signals can be improved upon by integrating them with macro-quantamental concepts.

For example, an FX forward-implied carry signal value can typically be enhanced by considering inflation differentials and cross-country differences in economic performance (view post here). Also, carry metrics can be enhanced by currency over- or undervaluation, using power parity-based valuation estimates that are partly or fully adjusted for historical gaps (view post here).

There are two simple ways to enhance carry strategies with economic information:

  • The first increases or reduces the carry signal depending on whether relevant economic indicators reinforce or contradict its direction. The output can be called “modified carry”. It is a gentle adjustment that leaves the basic characteristics of the original carry strategy intact.
  • The second method equalizes the influence of carry and economic indicators, thus diversifying over signals with complementary strengths. The combined signal can be called “balanced carry”. Empirical analysis suggests that both adjustments would have improved the performance of FX carry strategies (view post here).

Commodity carry signals have also profited from some quantamental enhancements. Carry on commodity futures contains information on implicit subsidies, such as convenience yields and hedging premia. Its precision as a trading signal improves when incorporating adjustments for inflation, seasonal effects, and volatility. There is strong evidence for the predictive power of various metrics of real carry with respect to subsequent future returns and stylized naïve PnLs based on real carry points to material economic value (view post here).

Equity allocation

Macroeconomic trends affect stocks differently, depending on their lines of business and their home markets. Hence, point-in-time macro indicators can support two types of investment decisions:

  • allocation across sectors within the same country and
  • allocation across countries within the same sector.

A number of plausible quantamental categories have proven relevant for predicting relative sectoral and relative country returns (view post here).

There is sound reason and evidence for the predictive power of macro indicators for relative sectoral equity returns. However, the relations between economic information and equity sector performance can be complex. Considering the broad range of available point-in-time macro-categories that are now available, statistical learning has become a compelling method for discovering macro predictors and supporting prudent and realistic backtests. For developed equity markets, a simple learning process produces signals that are positive predictors for the relative returns of all GICS sectors versus a broad basket. Combined into a single strategy, these signals create material and uncorrelated investor value through sectoral allocation alone (view post here).

Success in cross-country macro trading often relies on differentiating indicators related to monetary policy and corporate earnings growth in local currency. A straightforward, non-optimized composite score of quantamental indicators across countries could have added significant value to an equity index futures portfolio beyond simple risk-parity exposure. Furthermore, a purely relative value equity index futures strategy across countries would have produced respectable long-term returns (view post here).

Higher-frequency macro signals

The first differences of a macro-quantal time series are proxies of market information state changes, i.e., point-in-time updates of recorded economic developments. They can refer to a specific indicator or a broad development, such as growth or inflation. The broader the economic concept, the higher the frequency of changes. Information state changes are valuable trading indicators. They provide daily or weekly signals and naturally thrive in periods of underestimated escalatory economic change, adding a layer of tail risk protection.

For example, one can produce signals-based information state changes to interest rate swap trading across developed and emerging markets, focusing on growth, sentiment, labour markets, inflation, and financing conditions. Normalized information state changes are comparable across economic groups and countries and, hence, can be aggregated to local and global signals. The predictive power of aggregate information state changes has been strong, with material and consistent PnL generation (view post here).

Integration with statistical learning

For almost all types of financial contracts and positions, one can build a wide range of candidate macro-quantamental signals based on theory and plausibility. The challenge is to select from these candidates and combine them with a method that is consistent across time and allows realistic backtesting. This is where statistical learning and macro-quantamental signals become natural allies.

Statistical learning offers methods for sequential model selection, as well as associated hyperparameters, for signal generation, thus supporting realistic backtests and automated operation of strategies. It can be applied to sequential optimization of three important tasks: feature selection, return prediction, and market regime classification (view post here).

For example, regression-based statistical learning is a simple and easy-to-understand method for combining macro indicators into a single trading signal (view post here). It has proven its value for combining a range of macro factors into a single FX trading signal (view post here). Notably, signals based on regression coefficients can be adjusted for statistical precision to align intertemporal risk-taking with the predictive power of signals (view post here).

Resistance to the macro-quantamental expansion

The industry-wide availability of quantamental indicators has principally connected the worlds of macroeconomics and systematic trading. This is irreversible. However, penetration of this new technology affects the investment process at its core and, hence, clashes with institutional inertia and individual resistance:

  • Congested data pipelines: Information technology has allowed the production of a growing range of data sets, often called “alternative”. The investment management industry is a natural target of their monetization. However, asset managers face margin and cost pressures and need to filter the flood of quantitative information. They require time for exploration and adaptation. Macro-quantamental indicators may be new formats rather than new types of information. Still, they are classified as “new datasets” and, hence, often travel on a crowded procurement and exploration road.
  • Distrust towards macroeconomics: Producing investment value with macro-quantamental indicators works best with basic domain knowledge of macroeconomic statistics and economic theory. This is not popular with many systematic strategy developers who mostly hail from the realms of software engineering, data science, and mathematical finances. The distrust of macroeconomics is enhanced by the impenetrable jargon of economic research papers and the notorious tendency of economists to emphasize disagreements rather than consensus.
  • Exclusivity: A quantamental system is what economists would call a “club good”, i.e., a type of service that benefits from having multiple but controlled numbers of users, a bit like a tennis club or a swimming pool. For example, JPMaQS, the first global quantamental system, is mainly for participating institutional investors. While it is free for research, a subscription fee is required for complete service.
  • Misunderstandings of costs: A macro-quantamental “club” may charge higher fees than a standard economic database. However, tradable and non-tradable data are not comparable services, as the former includes modelling and refinement that otherwise would accrue as costs with the data user. Moreover, the cost of quantamental indicators relative to related trading profits is typically marginal. Indeed, for research purposes, up to the proof of concept of a quantamental signal, quantamental indicators are typically free. Compared to ad-hoc in-house development of quantamental indicators, access to a quantamental system usually implies significant cuts in development times and costs.
Share

Related articles