Leithner Letter: OECD Chair Warns “Our Entire System Is Unstable”

period around the 2008 housing crisis and recession.”⁶ Figure 3 plots Baker et al.’s global index of policy uncertainty.⁷ Since 1997, it’s averaged 106; since 2008, it’s averaged 135; and during November and December 2016 it exceeded 270.

That’s not just a record high: its 3.8 standard deviations above its mean since 1997 and 3.1 times its mean since 2008. From the point of view of global econom-ic policy, the four most “unsettled” months since 1997 all occurred in 2016 (June, July, November and December). In statistical terms, the present uncertainty of policy significantly exceeds its long-term average; in plain English, it’s now very – and perhaps perilously – high.

Around the world, ambiguity currently surrounds vital matters including:

1. America’s global role: will Mr Trump’s presidency generate a new order or intensify the pre-existing disorder? Can the U.S. Government continue to underpin the world’s economic and financial systems, and to underwrite the West’s military alliance? Even if it’s able, will it remain willing to do so? In order to “make America great again,” Mr Trump threatens trade wars (and continues the “currency war” that the Obama Administration commenced); will they beggar and destabilise other parts of the world?
2. China’s rise: will it continue to occur peacefully? Or will events in the South China Sea or elsewhere take unexpected and even hostile turns?
3. The European Union: what form – hard, soft or other – will “Brexit” take? How will it affect Britain and Europe? Will other countries, such as France and Holland, leave? Will debt crises push still others, such as Greece, out the door? Can – indeed, should – the EU survive?
4. Governments’ finances: when will budgets in Australia, the U.S. and else-where return to surplus, or at least to balance? Will economic growth pro-vide the traction or must cuts of expenditure do the heavy lifting?
5. Interest rates: will central banks continue to suppress rates of interest at his-torically-unprecedented lows? Or, as the Fed and others have frequently hinted but repeatedly failed to do, will they lift – albeit slightly – the foot from the accelerator? What happens to real estate, stock and other markets if and when they do?

Measuring Economic Policy Uncertainty and Financial Market Volatility

According to their website ( www.policyuncertainty.com), in order to measure “policy-related economic uncertainty” Scott Baker of Northwestern University, Nick Bloom (Stanford University) and Steven Davis (University of Chicago) construct an index from three types of underlying components. One component quantifies newspaper coverage of policy-related economic uncertainty. A second component reflects the number of federal tax code provisions set to expire in future years. The third component uses disagreement among economic forecasters as a proxy for uncertainty.⁸

The Volatility Index (VIX), which is known colloquially as “the fear gauge,” measures the implied volatility of “call” and “put” options. In plain English,

1. the higher are the premiums that buyers of “put” options are willing to pay (perhaps, among other things, because they fear that the prices of stocks they own will shortly fall), and/or
2. the higher are the premiums that buyers of “call” options are willing to pay (perhaps because they fear that the prices of stocks they don’t own will shortly rise),

the higher is the implied volatility of the market in question – and the greater are market participants’ fear and greed regarding the future. A VIX of 20 for the Standard & Poor’s 500 Index on a given day, for example, implies that buyers and sellers of options over stocks – and, by inference, buyers and sellers of stocks as a whole – expect a movement of 20%, up or down, during the next year.

Two Points about VIX

“The last time VIX started the year [as low] as it has in 2017,” writes James Mack-intosh (“What the VIX Is Really Telling the Markets Now,” The Wall Street Jour-nal, 13 February 2017), was in 2007, shortly before the subprime crisis hit. Before that, it was 1994, a year when the U.S. Federal Reserve shocked markets and hedge funds blew up. The historical parallels are scary because when inves-tors anticipate that volatility will be low, it can be sign of excessive complacency.

How well does VIX, a measure of implied volatility, anticipate actual volatility? Figure 4 plots monthly averages of VIX, expressed as percentages and net of the S&P 500 Index’s actual volatility (i.e., its maximum value less its minimum value, expressed as a percentage of its minimum value) during the subsequent twelve months. The good news is that, on average since 1990, VIX has anticipated sub-sequent volatility extremely well: its mean is 19.7% whereas the mean of actual (subsequent) volatility is 19.8%. Yet over extended periods VIX has predicted less well: during the first half of the 1990s, for example, it usually exceeded actual volatility; conversely, during the decade’s second half actual volatility typically exceeded implied volatility. The bad news: on the eve of the Global Financial Crisis, VIX was as blind as most other market participants. Throughout 2007and into 2008, it implied an up-or-down movement of the S&P during the next twelve months of ca. 20%; at its maximum, subsequent volatility exceeded 80%.

A second point, which Figure 5 illustrates, is that VIX has tended, when it ap-proaches extremes, subsequently to regress towards its mean. A very low (i.e., lowest decile of daily observations since 1990) observation today leads, on aver-age, to a higher one six months hence; and a comparatively one high today (i.e., highest decile) is associated with a lower one six months later. For the “middle” 80% of observations, VIX fluctuates close to its mean.

Uncertainty and Volatility in the U.S.

According to Baker et al.’s website,

current levels of economic policy uncertainty are at extremely elevated levels compared to recent history. Since 2008, economic policy uncer-tainty has averaged about twice the level of the previous 23 years. A significant dynamic relationship exists between our economic policy uncertainty index and real macroeconomic variables. We find that an increase in economic policy uncertainty as measured by our index fore-shadows a decline in economic growth and employment in the follow-ing months. … Moreover, since 2008, an increasingly large share of these large stock movements [has] been caused by policy-related events.

(adsbygoogle = window.adsbygoogle || []).push({});

@media (max-width: 600px) {
#admobil {
display:none;
}
#adpost {
display:inline-block;
width:300px !important;
height:250px;
}
}

@media (min-width: 600px) {
#admobil {
display:none;
}
#adpost {
display:inline-block;
width:336px;
height:280px;
}
}

Figure 6 plots Baker et al.’s measure of economic policy uncertainty in the U.S. since 1900.⁹ Before the Great Depression of the 1930s, it fluctuated but usually remained below its average (116) for the entire series. During the Depression, it rose sharply and to a level well above the long-term average; it also occasionally spiked to near-record levels (most notably, above 300 in early-1935, when the Supreme Court declared that key components of Franklin Roosevelt’s New Deal were unconstitutional). During the next quarter-century, uncertainty fell steadily and cumulatively drastically (i.e., to below 32 in October 1957 and again in September 1958); as a result, from the late-1940s until the late-1960s it remained well below its long-term average. Seldom since the early-1970s, in contrast, has it fall-en much or for long below this average. It rose steadily from the mid-1950 to the late-1980s; since then, it’s fluctuated at above-average levels and occasionally spiked sharply upwards. In