Is the sharp decline in the Index of Leading Economic Indicators over the past year a forecast of recession or a "false" signal? The debate over the economy's future course is far from academic, as it has significant consequences for policymakers, businesses, and investors.
Two index components, the yield curve and the ISM new orders diffusion index, account for most of the decline in the leading economic indicators over the past year. Both components' negative signals are questionable for fundamental and technical reasons. The weight of evidence supports the "false" signal scenario.
Leading Indicators As A Forecasting Tool
Victor Zarnowitz, a preeminent scholar on leading economic indicators and their use in forecasting business cycles, stated, "Some leading indicators turn out to be most operative and useful in one set of conditions, and others in a different set,"... and "there are a number of frequently observed regularities (e.g., inverted yield curve) that seem likely to persist and play important roles in business cycles but are certainly not immutable."
The yield curve signal has been "useful" and "immutable" as it predicted every recession since the 1950s. But the "set of conditions" that gave it greater accuracy than other indicators no longer exists, given the introduction and scale of the new Fed policy tool QE.
The primary purpose of QE is to keep or anchor long-term interest rates lower than otherwise, a factor not present in previous episodes of yield curve inversion. Massive asset holdings by the Fed reduce the yield curve's slope, creating the potential for curve inversion, thereby distorting what was previously a free-market-based yield curve signal. Uncertainty about the scale of the effect is high, but there is certainty about the impact, which is lower long-term bond yields.
The distortion in the yield curve will generally get smaller as the Fed balance sheet shrinks. Yet, the Fed's balance sheet is still too large ($7 trillion+, or more than $3 trillion above the level of 2020 and much more than that when QE was first introduced in 2012) to conclude with any confidence that yield curve inversion is not a "false" signal.
(Note: Another reason to fade the inverted yield curve recession signal is that the 10-year yield is well below the growth in Nominal GDP. There has never been a recession with the 10-year yield below Nominal GDP.)
The Institute of Supply Management (ISM) new orders index is the second-largest negative contributor in the past year. The new orders index is a diffusion index that captures the "breadth " of change and how many industries report gains or declines in new orders. Diffusion series have limitations or shortcomings.
The leading economic index includes two other new order series, the dollar-based measures of new orders for nondefense capital goods, excluding aircraft and consumer goods. Is that overkill, with three indicators that comprise 30% of the leading index covering the same thing: new order flow? Maybe.
Yet, what is perplexing is that the "hard" data series, the dollar-based orders series, and the."soft" data series, the orders diffusion index, are in sharp contrast. The dollar-based series have been positive in the past year, whereas the diffusion of new orders has been decidedly negative.
The economy runs on the number of dollars spent, so the dollar-based orders signal is consistent with GDP growth figures, while the diffusion index of orders is not.
Could technical issues, such as a drop-off in the number of purchasing managers' responses, impact the diffusion index signal? That is a possibility, but it is clear that the negative contribution derived solely from the diffusion index of new orders is flawed because it is inconsistent with what is happening.
The ISM new orders series and the current yield curve measure are recent additions to the leading economic index. The ISM order series became part of the index after the 2007-09 financial crisis and the current measure of the yield curve following the 2001 recession. Their inclusion means they replaced other indicators whose signal proved less robust or "useful."
However, their inclusion does not guarantee a robust or "useful" signal in the current cycle. The weight of evidence---comparing the economy's performance relative to both series signals---suggests their utility has diminished, if not disappeared. Thus, the "false" signal scenario is more plausible.
Investors should not expect the Conference Board, the global institution responsible for estimating the leading economic index, to suddenly say that the current composition of the leading index is flawed and needs to be revised. The staff needs time to research how well each series measures the economic variable or process in question and whether each series still exhibits the leading characteristics they did in prior cycles. Revisions in the leading index after the financial crisis took three years.
In the meantime, investors should give more attention to other leading indicators, such as stock prices and corporate profits, as they have been more in tune with what has been happening in the economy.
Comentários