Historical relationships between economic data play a key role in shaping expectations. In the US, the Sahm rule is such an important stylised fact: when the recent increase in the unemployment rate reaches a certain threshold, a recession tends to follow shortly or has even already begun. The jobs report published early August showed that this critical value had been reached, triggering a drop in investor sentiment. At the Jackson Hole conference, Jerome Powell explained that the Fed’s focus is shifting to the labour market and brought an unambiguous message that the rate cutting cycle is to start in September. In the Eurozone, the negative relationship between companies’ employment expectations and households’ unemployment expectations over the next 12 months is an important stylised fact, so it was a relief to see that in August the former increased after several months of decline, although the latter worsened slightly. Unemployment expectations and the inflation outlook are highly correlated with household expectations about their financial situation, which in turn influences consumer spending. As disinflation continues, the focus of analysts and the ECB should gradually shift to the labour market, like is already the case in the US.
August often sees a bout of increased market volatility. This was again the case this year, following the disappointing US labour market data and the surprise decision by the Bank of Japan to hike its policy rate as well as its guidance that more tightening will follow. Although markets recovered quite rapidly thereafter, the events -especially those in the US- are a reminder of the crucial interaction between facts -economic data-, stylised facts -the historical relationship between economic data and between these data and market developments- and sentiment. Bad economic news can trigger a drop in sentiment, which in turn could cause future data to show more weakness if historically a close link has been established between such data and the economy in general. It illustrates the risk of a self-reinforcing negative spiral (self-fulfilling pessimism as described by Susan Collins, Boston Federal Reserve president[1]).
Early August, the observation that the US unemployment rate had increased for the fourth month in a row, reaching 4.3%, had raised concerns about a looming recession risk because the threshold of the ‘Sahm rule’ had been broken.[2] This ‘rule’ is a textbook example of a stylised fact. Historically, when the unemployment rate has moved above the threshold, it marks the start of a recession. However, one should not see this as a foolproof signal. Rather, it forces us to look even more diligently to any evidence that would point in this direction. This is also what the Federal Reserve is doing. Speaking at the Jackson Hole conference, Fed Chair Jerome Powell insisted that “the economy continues to grow at a solid pace. But the inflation and labor market data show an evolving situation. The upside risks to inflation have diminished. And the downside risks to employment have increased.”[3] It was an unambiguous message that the rate cutting cycle is to start at the FOMC’s meeting on 17-18 September, with the pace of further reductions dependent on the data and especially the labour market data. The Eurozone also has its stylised facts. An important one at the current juncture is the relationship between the employment expectations of companies and households’ unemployment expectations over the next 12 months. As shown in Chart 1, when the former decline, the latter tend to increase. Although unemployment expectations are not part of the European Commission’s consumer confidence index[4], there could still be an influence on household spending.
EUROZONE: CONSUMERS’ ASSESSMENT OF THE FINANCIAL SITUATION OVER THE NEXT 12 MONTHS: DRIVERSTable 1 shows the relationship between several consumer confidence measures and quarterly growth of real household consumption.[5] The financial situation over the next 12 months offers the best result, closely followed by the Commission’s confidence index. Unemployment expectations show the lowest statistical fit.[6]
EUROZONE: REAL HOUSEHOLD SPENDING AND CONSUMER CONFIDENCEThese results, which show the central role of expectations about the financial situation, raise the question of possible influencing factors. On theoretical grounds, inflation expectations and unemployment expectations may play a role.[7] These can be captured by data from the European Commission’s consumer confidence survey, respectively the price trends over next 12 months and the unemployment expectations over the next 12 months. Because the relationship between these data and financial expectations may fluctuate, the data sample has been split in smaller subsamples, rather than using all data in a single regression. The results are shown in Chart 2. Interestingly, they fluctuate a lot in terms of the overall relationship (R², shown by the dotted line) between financial expectations and expectations about unemployment and inflation. During the most recent period, it has been high.
EUROZONE: COMPANIES’ EMPLOYMENT EXPECTATIONS AND HOUSEHOLDS’ UNEMPLOYMENT EXPECTATIONS OVER THE NEXT 12 MONTHSMoreover, the statistical significance (the t statistic, shown by the blue and green line) of the two explanatory variables fluctuates as well. It is good to keep this in mind when reflecting on how financial expectations may evolve in the near term. Based on the latest regression, expectations about inflation and unemployment are equally important. In a scenario of further disinflation, one would expect that expectations about price trends will play less of a role in shaping the financial expectations of households. This would mean that the focus of analysts and the ECB should gradually shift to the labour market, like is already the case in the US. Although in August, households’ unemployment expectations worsened slightly, the fact that companies’ employment expectations increased after several months of decline and is now close to its long-term average is providing some comfort.
[6] The low R² reminds us that a large part of the fluctuations in consumer spending depend on other factors than consumer confidence measures.
[7] Several other factors -such as interest rates, uncertainty, income, etc.- can also influence financial expectations. For a detailed econometric analysis of the drivers of consumer confidence in general, see a recent paper of the European Commission: Low consumer confidence and the economy. - Insights from the euro area, European Business Cycle Indicators, Technical Paper 074, July 2024.
