Data • Psychology • Technology in Perfect Alignment
The Internet of Behaviors draws on every available data stream to model and predict human decisions — but some of the richest signals sit not in app logs or clickstreams, but in the macroeconomic indicators that governments publish each month. Central banks watch them obsessively. Markets reprice within seconds of a new data release. And for anyone trying to understand why consumer confidence, spending patterns, and business investment shift the way they do, learning to read macro signals is a fundamental skill.
One of the most closely watched signals in the bond market is the yield-curve inversion signal. Under normal conditions, longer-dated government bonds pay higher yields than short-dated ones — lenders expect more compensation for tying up capital over a longer horizon. When that relationship flips, and short rates rise above long rates, it historically signals that investors expect economic weakness ahead. A deeply inverted curve preceded most of the recessions of the past half-century. For IoB practitioners, this matters because shifts in perceived economic safety ripple directly into consumer behavior: people defer large purchases, change savings habits, and reduce discretionary spending in ways that show up in transactional and behavioral data well before official GDP numbers confirm a slowdown.
The labor market adds its own layer of complexity. Most people know the unemployment rate, but how many people are actually working or looking for work — the labor force participation rate — is arguably more revealing. An unemployment rate can fall simply because discouraged workers stop searching; participation tells you whether people are genuinely engaged with the labor market. When participation is low and wages are rising anyway, it often signals genuine tightness rather than statistical artefact. Behavioral economists studying work-search patterns have found that participation responds to more than just wages: childcare availability, regional mobility, and even perceived job quality all feed into the decision to stay in or re-enter the labor force.
Closely related is the question of how fast workers expect pay to rise. Expectations matter because they are self-fulfilling: workers who believe wages will rise are more likely to accept job offers, while employers who expect competition will raise their bids pre-emptively. The Federal Reserve surveys household wage-growth expectations specifically because those beliefs shape spending and saving decisions today. When expectations become "unanchored" — drifting well above the central bank's inflation target — the resulting behavioral shifts can force aggressive monetary intervention.
Wage growth, of course, is meaningful only in relation to what workers actually produce. Output produced per hour worked — labor productivity — determines whether wage increases can be absorbed without triggering inflation. Rising labor productivity means each dollar of wages corresponds to more real output, making higher pay sustainable. When productivity stagnates while wages surge, profit margins are squeezed and price pressures build. The yield curve and productivity statistics thus form a natural pair: the yield curve signals where the economy is heading, while productivity data explains whether the economy can sustain the current trajectory without overheating.
Underlying all of these metrics is the question of money itself. The M2 money supply measures the total stock of money readily available in the economy: cash, checking deposits, savings accounts, and small-denomination time deposits. When M2 expands rapidly — as it did during pandemic stimulus — it typically precedes a rise in consumer spending, asset prices, and eventually inflation. When M2 contracts or grows slowly, the reverse tends to follow. Tracking M2 alongside the yield curve inversion signal gives a fuller picture: the curve tells you about market expectations for rates, while M2 tells you about the actual fuel available for spending.
For IoB systems that aim to predict shifts in consumer and business behavior, these five indicators — the yield curve, labor participation, wage expectations, labor productivity, and the M2 measure — are among the most powerful inputs available outside transactional data itself. They are publicly available, regularly updated, and richly studied. Building them into behavioral models alongside device-level and social-media signals is one of the underexploited opportunities at the frontier of the field.