Given that it is forecasting time again for next year’s budget, banks often use a variety of economic indicators to help forecast demand for credit, liquidity, and inflation. Often time, we will see many of these indicators in ALCO reports or strategic plans. We have tried a great many indicators and have tracked the effectiveness of each one. Today, we will cover some of the more unreliable ones, while tomorrow we will cover the ones that work.
Non-farm Payrolls: This is one of most influential data sets in the market and has one of the largest impacts on intermediate-term rates and equity prices. However, this Bureau of Labor Statistic number lags so much and has so many major revisions (three) that for a long-term predictor, it has a very low correlation to most factors that the bank wants to forecast. The most accurate use for this data set is to use the 12-week moving average as an indicator of credit quality, but even that is suspect as the correlation is less than 0.35 (1 being perfectly correlated).
Baltic Dry Index: This has been a favorite of ours for years as we thought that demand for shipping should be the first thing that picks up or slows down. While conceptually that’s true, after years of trying to correlate this index to growth (it explains less than 20% of growth), we have come to realize that it is practically worthless. The supply of ships has such a large impact on cargo prices, and the speculative nature of commodities skews this index so that it is an unreliable predictor of production, inflation or anything else that we have tested.
Consumer Confidence: We are not sure what consumers are feeling when they are asked about current conditions, but however they answer, consumers actions do not follow their words or people change 10 minutes after answering the questions. This index is more a measure of sentiment than actual future economic activity and is unreliable when it comes to predicting loan growth, credit, rates or anything else.
Durable Goods: Like Confidence and Non-farm Payroll, this is another indicator that moves markets, but has very little long-term predictive power. That said, the year-over-year change in the 3-month moving average of core CAPEX (investment), which is a component of the Durable Goods number, is somewhat correlated to corporate cash levels. The more investment, usually the lower balances in operating accounts are predicted.
The Volatility Index: The Volatility Index or VIX (also called the “fear index”) is synthetic index of change or equity market optionality. As a general rule, the VIX spikes in periods of uncertainty. While normally averaging around 14, the index has been trending up and closed yesterday at 18.7. Lately, it has had a couple days above 20 which is a sign of danger. What does all this mean? Not much for banks. It is fun to talk about in Board and Risk meetings, but since it is a measure of equity movement, it is really not helpful for forecasting interest rates, credit, loan/deposit optionality or anything else you might want to project.
We have only highlighted the most common economic indicators that are fairly worthless. Other, less popular ones that are equally suspect include the S&P Agricultural and Livestock index, used to predict agriculture demand or credit quality (less than a 0.55 correlation to either) and crude oil prices (0.60 correlation to production inflation). Fret not, as tomorrow we will highlight some key indicators that do have some predictive power and can help your bank more accurately forecast the future.
Submitted by Chris Nichols on October 16, 2013