Home » Research Blog » Macro trading and macroeconomic trend indicators

Macro trading and macroeconomic trend indicators

Macroeconomic trends are powerful asset return factors because they affect risk aversion and risk-neutral valuations of securities at the same time. The influence of macroeconomics appears to be strongest over longer horizons. A macro trend indicator can be defined as an updatable time series that represents a meaningful economic trend and that can be mapped to the performance of tradable assets or derivatives positions. It can be based on three complementary types of information: economic data, financial market data, and expert judgment. Economic data establish a direct link between investment and economic reality, market data inform on the state of financial markets and economic trends that are not (yet) incorporated in economic data, and expert judgment is critical for formulating stable theories and choosing the right data sets.

The below is an updated version of this site’s summary on systematic trading strategies based on macro trends.

Why macroeconomic trends matter

Macroeconomic trends move asset prices for two reasons. They influence investors’ attitudes towards risk and they affect the risk-neutral expected payoff of securities. An example of the first is the rise in risk aversion in economic recessions when cash flows and incomes fall to critical thresholds. Examples of the second are the impact of inflation on the real return on nominal fixed income securities, the influence of economic growth and relative price-wage trends on the earning prospects of stocks, the effect of financial conditions on the default risk of credit, and the relation between external balances and exchange rate dynamics.

Due to the pervasive influence of macroeconomic trends, most investors watch economic data releases and employ economists to analyze them. Empirical studies show that bond and equity markets are more likely to post large moves on days of key data releases than on other days (view post here). However, the influence of economic data on market price changes is the stronger the longer the time horizon that we consider. This is because economic changes are typically more persistent than non-fundamental factors. They are therefore a major explanatory variable of medium to long-term price trends.

  • For the fixed-income market, it has been estimated that on a quarterly basis more than a third of bond price fluctuations in the U.S. can be explained by deviations in the country’s major published economic data from analyst expectations (view post here). By contrast, data surprises explain only 10% on market fluctuations on a daily basis. Medium-term returns of government bonds seem to be predictable through nowcasted economic growth, as well as measures of financial market tail risk (view post here).
    Over the longer-term bond yields seem to move almost one-to-one with expected inflation and the estimated equilibrium short-term real interest rate (view post here). Moreover, research claims that most of the decline in equilibrium real interest rates from the 1980s to 2010s may be explained by a single fundamental divergence. On the one hand, the propensity to save surged due to demographic changes (view post here), rising inequality of wealth and the reserve accumulation of emerging market central banks. On the other hand, investment spending was held back by cheapening capital goods and declining government activity (view post here).
  • In the foreign exchange space, both theory and evidence point to a close relationship between relative business cycles and exchange rate dynamics (view post here). Currencies of countries in a strong cyclical position should appreciate against those in a weak position. Also, deviations of currency values from their medium-term equilibrium give rise to multi-year exchange rate trends. Indeed, long-term empirical evidence for developed market currencies suggests that real exchange rates have been mean-reverting and that re-alignment occurs mainly through the nominal exchange rate (view post here).
  • As to equity, theoretical and empirical research suggests that a downshift in expected inflation raises average company valuation ratios, such as price-earnings ratios, and credit default risk at the same time, thus giving rise to a relative asset class trend (view post here). Also, some economic estimates suggest that all of the real stock market gains in the U.S. since the 1980s are caused by the gradual redistribution of the benefits of productivity gains from workers to shareholders (view post here).
  • In commodity markets, the big cycles in some raw material prices have been driven mainly by “demand shocks”, which seem to be related to global macroeconomic changes and have persistent effects of 10 years or more (view post here).

Economic shocks have more powerful market effects if they change long-term expectations. Thus, a key factor of economic impact is whether long-term expectations are “anchored” or not. For example, persistent undershooting of inflation targets in the developed world has made long-term inflation expectations more dubious and susceptible to short-term inflation trends. This “de-anchoring” can be measured (view post here) through surveys and long-dated securities, providing valuable information on the consequences of price shocks for markets.

How to align macroeconomic trends with market positions

Often enough the directional effect of economic change is straightforward, following standard macroeconomic theory and market experience. For example, rising expected inflation and lower unemployment have historically translated into higher low-risk bond yields (view post here).  Also, swings of large commodity-intensive sectors, such as construction in China, have driven global prices for raw materials, such as base metals (view post here). Furthermore, export price changes of “commodity countries” help to explain and even to predict their exchange rate dynamics (view post here).

However, macroeconomic trends can also have multiple effects, which need to be disentangled. For example, expansionary financial conditions can be both beneficial and harmful for future equity market performance, depending on the trade-off between positive growth impact and elevated vulnerability. On these occasions, indicators need to be modified, become parts of larger formulas, be split into different parts. For example, financial conditions can be divided into short-term impulses, such as yield compression, and medium-term vulnerability, such as increased leverage (view post here). Combinations of negative shocks and elevated vulnerability would then be clear negative signals for equity markets. Combinations of positive impulses and low vulnerability would be clear positive signals.

The relation between macroeconomic trends and financial returns is also often obfuscated by global factors. For example, the value of a currency typically benefits from a strengthening of the underlying economy relative to other countries. However, almost all currencies are to varying degrees sensitive to changes in global markets and the exchange rates of the largest economies. In order to validate and trade relative economic trends it therefore useful to hedge against such global influences or set up positions relative to similar contracts or both. Empirical evidence suggests that global FX forwards can be hedged reliably against the largest part of global market influences (view post here).

Sometimes information regarding economic uncertainty can be as valuable as information on economic direction. One can estimate economic uncertainty through various methods, such as keyword frequency in the news, relevant market volatility and forecast dispersions. Such measures help to detect phases of popular fear or panic and complacency (view post here), both of which offer opportunities for professional investors. Indeed, composite measures suggest that uncertainty typically rises abruptly but subsides only gradually.
Unsurprisingly, uncertainty about the economic and financial state in general has been conducive to higher volatility in market prices, including commodities (view post here). Economic uncertainty can also affect directional trends. For example there is evidence that uncertainty about external balances leads to underperformance of currencies of economies with net capital imports (view post here).

Macro trend indicators

Since the range of available macro data is vast they must inevitably be condensed into small manageable sets of meaningful indicators. Generally, a macro trend indicator can be defined as an updatable time series that represents a meaningful economic or financial trend and that can be mapped to the performance of tradable assets or derivatives positions. There are three major sources of information for macro trend indicators:

  • economic data,
  • financial market data, and
  • expert judgment.

While these sources are often portrayed as competing investment principles, they are highly complementary. Economic data establish a direct link between investment and economic reality, market data inform on the state of financial markets and economic trends that are not (yet) incorporated in economic data, and expert judgment is critical for formulating stable theories and choosing the right data.

Economic data

For all major economies, statistics offices publish wide arrays of economic data series, often with changing definitions, elaborate adjustments, multiple revisions and occasional large distortions. Monitoring economic data consistently is tedious and expensive. Most professional investors find it easier to trade on data surprises than on actual macro trends. It is not uncommon for investment managers to consider an economic report only with respect to its presumed effect on other investors’ expectations and positions and to subsequently forget its contents within hours of its release.

What makes monitoring economies difficult is that there is usually no single series that represents a broad macroeconomic trend on its own in a timely and consistent fashion. To begin with, conventional economic data are published with considerable lags, subject to frequent revisions, and often their true history is very hard to reconstruct for financial market backtesting. Moreover, many important types of macro information for markets are not produced by central agencies. For example, equilibrium real interest rates and long-term inflation trends are essential factors for fixed income strategies (view post here). Yet neither of these is available as an official reliable data series since such estimation requires strong judgment and macroeconomic modelling (view post here).  Even something apparently simple such as an inflation trend demands watching many different data series at the same time, such as consumer price growth, “core” inflation measures, price surveys, wage increases, labour market conditions, household spending, exchange rates and inflation derivatives in financial markets. In practice, the use of economic data for macro trading requires [1] producing special tradable economic data, [2] formulating a plausible and logical theory to create meaningful indicators, and [3] applying statistical methods.

Published economic data cannot be easily and directly plugged into systematic trading strategies. Unlike financial market data, which are intensively used for algorithmic and systematic trading, economic data come with a number of inconvenient features such as low frequency of updating, lack of point-in-time recording and backward revisions. Therefore, economics statistics and other quantifiable information must be brought into a form that is suitable for systematic research. One can call this form tradable economic data (view post here).

Theoretical structure establishes a plausible relation between the observed data and the conceived macroeconomic trend. This is opposite to data mining and requires that we set out a formula based on our understanding of the data and the economy before we explore the actual data.

  • As a most simple example, different sectoral production reports can be combined by adding them in accordance with the weight of the sectors in the economy.
  • The monetary policy stance in a regime with sizeable asset purchase programs can be estimated as a single “implied” short-term interest rates based on the actual short-term interest rate and the equivalent effect of compression of term premia, based on a yield curve factor model (view post here).
  • As a more advanced example, we can extend measures of consumer price inflation by indicators of concurrent aggregate demand. This helps to distinguish between supply and demand shocks to prices, making it easier to judge whether a price pressure will last or not (view post here).
  • Even modern academic macroeconomic theory can help. True, dynamic stochastic general equilibrium models are often too complex and ambiguous for practical insights. However, simplified static models of the New Keynesian type incorporate important features of dynamic models, while still allowing us to analyze the effect of macro shocks on interest rates, exchange rates and asset prices in simple diagrams (view post here for interest rates and here for exchange rates).

Statistical methods become useful where our prior knowledge of data structure ends. They necessarily rely on the available data sample. In respect to economic trends, they can accomplish two major goals: dimension reduction and nowcasting.

  • Dimension reduction condenses the information content of a multitude of data series into a small manageable set of factors or functions. This reduction is important for forecasting with macro variables because many data series have only limited and highly correlated information content. (view post here).
  • Nowcasting tracks a meaningful macroeconomic trend in a timely and consistent fashion. An important challenge for macro trend indicators is timeliness. Unlike financial market data, economic series have monthly or quarterly frequency, giving only 4-12 observations per year. For example, GDP growth, the broadest measure of economic activity, is typically only published quarterly with one to three months delay. Hence, it is necessary to integrate lower and higher-frequency indicators and to make use of data releases with different time lags.

In recent years, dynamic factor models have become a popular method for both dimension reduction and nowcasting. Dynamic factor models extract the communal underlying factor behind timely economic reports and translate the information of many data series into a single underlying trend (view post here and here). This single underlying trend is then interpreted conceptually, for example as “broad economic growth” or “inflation expectations”. Also, financial conditions of an economy can be estimated by using dynamic factor models that distil a broad array of financial variables (view post here).

It is important to measure local macroeconomic trends with a global perspective. Just looking at domestic indicators is almost never appropriate in an integrated global economy. As a simple example, inflation trends have increasingly become a global phenomenon, as a consequence of globalization and convergent monetary policy regimes. Over the past three decades, local inflation has typically been drifting towards global trends in the wake of deviations (view post here). As an example of the global effects of small-country shocks, “capital flow deflection” is a useful conceptual factor for emerging markets that stipulates that one country’s capital inflow restrictions are likely to increase the inflows into other similar countries (view post here). In order to measure this effect, one needs to build a time series of capital controls in all major economies in order to distil the specific impact on a single currency.

Financial market data

Financial market data are available faster and at higher frequencies than macro data. Also, investment professionals often find them easier to understand. However, extracting specific macro trend information content from financial data can be challenging as a single price typically reflects the influence of many factors. Hence, as with economic data, it takes theoretical modelling and statistical methods to translate market prices into macro factors.

  • A simple example would be to derive inflation expectations from breakeven inflation, as priced in the inflation swap markets. For this purpose, we must at least make an adjustment for the inflation risk premium embedded in the swaps contract, possibly by using the correlation of inflation swaps with broad market benchmarks (view post here).
  • Trends in industrial commodity prices are typically aligned with global demand, economic growth and, ultimately, inflationary pressure. Since commodity prices are observable in real-time they can predict related economic trends. And since most of these economic trends matter for interest rates, they help to forecast bond returns (view post here). More advanced information extraction would be to check whether rising commodity prices have coincided with upward or downward revisions to global industrial activity. This helps to distinguish between commodity supply and global demand shocks; these two can have very different implications for exchange rate, equity and rates markets (view post here).
  • Another intuitive source of information on perceived uncertainty is the futures curve of implied equity index volatility, particularly VIX (view post here). This curve is shaped by the relation between present and future expected volatility and, hence, serves as an indicator of present complacency, in form of a steep curve, or panic, through an inverted curve.
  • Bond and swap yields are a rich source of information. The level of real short-term rates is related to the monetary policy stance. The slope of the curve is related to expected future policy rates and risk premia. And the curvature of the term structure is naturally related to the expected “over-tightening” or “under-tightening”of monetary policy and, hence, is a valid trading signal for the foreign exchange market (view post here). Moreover, the difference between government bond yields and swap yields, adjusted for credit risk, is often indicative of a “liquidity yield” or “convenience yield” of government bonds, i.e. non-pecuniary benefits that arise from high liquidity, suitability as collateral and eligibility as regulatory liquidity buffers. Such liquidity yields not only indicate long-term expected returns of the bonds themselves, but their changes also affect exchange rate dynamics in a similar manner as changes in interest rates (view post here). For example, since the dollar exchange rate clears the market for safe dollar assets increases in the convenience yield for these assets typically trigger an overshooting in the international value of the dollar (forthcoming post).
  • Another simple and popular example is the measurement of monetary policy uncertainty through short-term rate derivatives. Policy uncertainty is a key component of equity return volatility that improves predictions that are otherwise based on historical and implied equity volatility alone (view post here).
  • The term premia in credit default swap curves are indicative of country financial risk. In particular, flattening or inversion of CDS curves is typically indicative of negative country-specific shocks (view post here). Empirical research suggests that changes in CDS term premia has predicted exchange rate changes and local stock returns in the past.
  • The USD exchange rate has become an important early indicator for U.S. and global credit conditions (view post here). This is because a large share of corporate loans is regularly sold on to mutual funds. In times of USD strength credit funds typically experience outflows, as the balance sheets of non-U.S. borrowers deteriorate, i.e. the weight of their USD debt increases relative to non-USD assets.
  • Equity and bond market volatility can be decomposed into persistent and transitory components by means of statistical methods. Plausibility and empirical research suggest that the persistent component of price volatility is associated with macroeconomic fundamentals. This means that persistent volatility is an important signal itself and its sustainability depends on macroeconomic trends and events (view post here). Meanwhile, the transitory component, if correctly identified, is more closely associated with market sentiment and can indicate mean-reverting price dynamics.
  • An example that relies more on statistical estimation would be the measurement non-conventional monetary policy shocks based on asset prices. For this we can estimate changes in the first principal component of bond yields that are independent of policy rates and on monetary policy announcement dates. Non-conventional monetary policy shocks tend to have a profound and lasting impact on most asset markets (view post here).

A global perspective is even more important for financial data than for economic data. In particular, U.S. financial markets have worldwide influence. Empirical research shows that shocks to U.S. monetary policy have a significant impact not only on USD exchange rates, but on foreign-currency risk premia more generally (view post here). Similarly, there is evidence that shocks to the term premium in longer-dated U.S. yields have a persistent subsequent impact on term premia in most other global markets (view post here).

Also, a cross-asset class perspective is important. Markets are still segmented insofar as different institutions and managers specialize in different types of information and assets. This is a form of rational inattention. For example, equity investors naturally focus more on corporate earnings prospects, while fixed-income investors pay more attention to macroeconomic trends and monetary policy. As a consequence, investment strategies in one market can often benefit from the information provided by another, if one is familiar with “decoding” price signals quickly. Thus, equity markets have historically been more sluggish than bond markets in adjusting discount factors to shifts in relative country inflation (view post here). Similarly, changes in the implied pace of future policy rates, as priced by fed funds futures, have in the past helped to predict equity returns (view post here) and even the U.S. dollar exchange rate (view post here).

Expert judgment

As a rule, expert judgment is a powerful complement rather than an alternative to statistical methods:

  • Experts can explain the meaning of data. Data analysis requires a good understanding of what the data really represent as opposed to what the label says. For example, some business surveys that refer to a particular month actually use data collected in the previous month. We also need expert judgment on the relevance of data. For example, in some countries, core inflation (excluding food and energy) is a very important benchmark for policy rates, while in other countries the central bank would only look at headline inflation.
  • Experts also help to detect data distortions. Data analysis needs timely and regular information on distortions, such as the impact of taxes or regulated prices on inflation statistics or the effect of natural disasters or calendar effects on growth.

Related articles