What markets tell us about macroeconomics

This article presents an analytical model1 to identify the macroeconomic factors that drive financial markets in the euro area and the US. It does so on the basis of correlations between different asset classes and daily high-frequency market data, spanning from June 2005 to the present.

For each region, we examine daily data from different types of financial assets: interest rates (overnight rates and sovereign bonds), inflation expectations (inflation swaps), equities (stock and volatility indices), exchange rates and commodities (gold, oil and gas).2

We then analyse the co-movements between these assets and extract patterns (in the form of common factors) that allow us to identify different macroeconomic drivers of market movements.3

Our exercise identifies three major categories of macroeconomic shocks: demand, supply and monetary policy. Intuitively, an increase in demand stimulates both economic activity and price growth, which is reflected in rising stock market prices (higher corporate profits), inflation expectations and higher interest rates (both to compensate for inflation and due to the possibility of central bank rate hikes). Conversely, a contraction in supply depresses activity and raises prices (this may be caused by an increase in the cost of commodities, such as oil), resulting in stock market declines and higher interest rates (amid expectations of monetary tightening to combat inflation).

Finally, a dovish monetary policy shock (e.g. when a central bank signals it will cut interest rates) is associated with declines in sovereign rates, a depreciation of the domestic currency, and an increase in stock markets and inflation expectations (due to the support of monetary easing for economic activity).

Formally, these three macroeconomic market drivers are identified through the various expected co-movements between financial variables, as summarised in the first table.4

Although it falls outside the scope of this article, our model reveals seemingly technical details which nevertheless provide «narrative» insights worthy of discussion.

Firstly, the most notable reading is the difference between the monetary policy shocks in the US and in the euro area.5 While the movements in both regions are in the expected direction, the exercise identifies different sensitivities and magnitudes between the two regions. In the euro area, equities are relatively less sensitive, generally speaking, to monetary policy shocks. This may reflect the greater weight of the financial sector in European stock market indices (a sector which tends to benefit from higher interest rates, unlike the rest). Another similar observation is the low sensitivity of inflation expectations to monetary shocks identified by the model in the US – a result consistent with the academic literature.6

Secondly, the analysis makes it possible to calibrate the sensitivity of each asset class to different types of shock. For example, equities show greater sensitivity in demand shocks than in supply shocks. In the case of fixed-income securities, sensitivity varies by region: in the euro area, it shows a similar magnitude in response to supply and demand shocks, whereas in the US it is greater in response to demand shocks.

Finally, it is difficult to decouple the movement of commodities from that of other assets using only the three drivers described. If we expand the number of drivers analysed, we obtain components with less aggregate explanatory power but which offer greater precision in explaining specific co-movements between assets. This allows us to observe how fluctuations in gold prices are heavily influenced by drivers that were of limited aggregate importance in the past, but which have recently surged.7