What Does Disagreement Tell Us about Uncertainty?

January 7th, 2015
in econ_news

by Matthew Ploenzke, Robert Rich, and Joseph Tracy - Liberty Street Economics, Federal Reserve Bank of New York

Uncertainty is of considerable interest for understanding the behavior of individuals as well as the movements in key macroeconomic and financial variables. Despite its importance, direct measures of uncertainty aren’t widely available. Because of this data limitation, a common practice is to use survey-based measures of forecast dispersion—reflecting disagreement among respondents—to proxy for uncertainty. Is this a reliable practice? Here, we review the distinction between disagreement and uncertainty as concepts, and show that this conceptual distinction carries over to their empirical counterparts, suggesting that disagreement is not generally a good proxy for uncertainty.

Follow up:


The terms “disagreement” and “uncertainty” refer to very different concepts. Disagreement refers to a collection of forecasts or point predictions and the nature of their clustering around each other—the more disperse the forecasts, the greater the extent of disagreement among the survey respondents. On the other hand, uncertainty refers to the distribution of the probabilities that a respondent attaches to the different possible outcomes of the forecasted variable—the more confidence held by a respondent, the tighter this distribution is and the lower the respondent’s uncertainty.

Although disagreement and uncertainty are different concepts, some commentators have drawn a connection between the two. That is, episodes characterized by high (low) disagreement are viewed as indicative of high (low) uncertainty shared by the respondents. This assumption provides the basis to use disagreement as a proxy for uncertainty when measures of the latter magnitude aren’t available.

Is the assumed positive association between disagreement and uncertainty plausible? Yes. Is it necessarily true? No. To understand why either case is possible, we can look at the following figure previously discussed in this paper by Victor Zarnowitz and Louis Lambros:

Zarnowitz Lambrose

The illustration on the left shows the forecasts and associated probability distributions of two hypothetical survey respondents—respondent A and respondent B. The close proximity of the forecasts (ŷA and ŷB) indicates low disagreement, while the tight distributions around each forecast indicate low uncertainty. However, the illustration on the right depicts another possible situation. Here the forecasts are unchanged, so disagreement remains low. But the probability distributions around each forecast are now much wider, indicating high uncertainty. The figure is important for two reasons. First, it bears directly on the question of the reliability of disagreement as a proxy for uncertainty. While the conditions depicted in the left illustration might justify this practice, the conditions depicted in the right illustration would not. Second, it shows that the dispersion of forecasts by itself is not necessarily informative about the level of uncertainty across respondents. Because most surveys only report the respondents’ forecasts, observing the degree to which the forecasts cluster together can’t tell you whether the illustration on the left or on the right in the figure is the more relevant situation.

So, is it possible to say something more definitive about which of the two illustrations may be a better description of the relationship between disagreement and uncertainty? The answer is yes, because there are a few surveys that have the unique feature of asking respondents to report both a forecast and their uncertainty—the latter being represented by a histogram in which respondents assign probabilities to pre-specified bins that provide a range of possible outcomes for the forecasted variable. The best-known of these surveys are the U.S. Survey of Professional Forecasters (US-SPF), the European Central Bank Survey of Professional Forecasters (ECB-SPF), and the Bank of England Survey of External Forecasters (BOE-SEF).

There are various techniques, such as calculating the variance/standard deviation or interquartile range, that can be used with these histograms to derive an estimate of each respondent’s uncertainty, which can then be used to construct a measure of aggregate uncertainty. The degree of disagreement across the respondents’ forecasts can be measured in the same way. If disagreement is a useful proxy for uncertainty, then the two variables should display a meaningful positive association.

As it turns out, there are several studies that have used data from the aforementioned surveys to investigate the relationship between disagreement and uncertainty. These studies include work byBoero, Smith, and Wallis (2008) for the BOE-SEF; Rich and Tracy (2010)for the US-SPF; and Rich, Song, and Tracy (2012) for the ECB-SPF. The conclusions from the analyses are quite similar—disagreement is a poor proxy for uncertainty.

To help understand this finding, the following graph uses ECB-SPF data from the first quarter of 1999 through the fourth quarter of 2013 and plots measures of disagreement and uncertainty for one-year-ahead inflation forecasts. Details on the construction of the variables are provided in the Rich, Song, and Tracy paper cited above.

Measures of Disagreement and Uncertainty in 1-Year-Ahead Inflation Forecasts

The average uncertainty measure shows a clear upward drift since the second quarter of 2007, denoted by the vertical line, and remains nearly 50 percent higher today than in the pre-crisis period. The disagreement measure shows a noticeable increase starting in 2008, with some fairly dramatic spikes evident during the global financial crisis. More recently, and in contrast to uncertainty, disagreement has declined and returned to levels more consistent with the range observed over most of the pre-2007 period. When we examine the series more formally, the correlation is essentially zero. If we limit the analysis to data through the second quarter of 2007 to avoid any concerns that the results may be unduly influenced by the global financial crisis, then the correlation becomes negative. There is a relatively straightforward explanation for the weak relationship—the uncertainty measure displays relatively smooth behavior while disagreement is more volatile. These features of disagreement and uncertainty depicted in the graph are very representative of the relationship for other variables, such as growth and unemployment, and other forecast horizons. In particular, while most of the other observed correlations for the ECB-SPF data were higher than zero, they were still quite low and could generally be described as weak.

Our discussion focuses on the question posed in the title of our post: what does disagreement tell us about uncertainty? While we recognize that disagreement could be strongly and positively related to uncertainty, and that it has been common practice in many analyses to assume such a relationship, we find that there is little empirical evidence to support this view. A better approach is to use measures of uncertainty when they are available, as well as to design more surveys that would have the capability to directly measure respondents’ uncertainty.

Disclaimer

The views expressed in this post are those of the authors and do not necessarily reflect the position of the Federal Reserve Bank of New York or the Federal Reserve System. Any errors or omissions are the responsibility of the authors.

Source: http://libertystreeteconomics.newyorkfed.org/2015/01/what-does-disagreement-tell-us-about-uncertainty.html#.VK0p7ivF9K4


About the Authors

Ploenzke_Matthew Matthew Ploenzke is a research associate in the Federal Reserve Bank of New York’s Research and Statistics Group.

Rich_robert Robert Rich is an assistant vice president in the Research and Statistics Group.

Tracy_joseph Joseph Tracy is an executive vice president and senior advisor to the Bank president at the Federal Reserve Bank of New York.









Make a Comment

Econintersect wants your comments, data and opinion on the articles posted.  As the internet is a "war zone" of trolls, hackers and spammers - Econintersect must balance its defences against ease of commenting.  We have joined with Livefyre to manage our comment streams.

To comment, just click the "Sign In" button at the top-left corner of the comment box below. You can create a commenting account using your favorite social network such as Twitter, Facebook, Google+, LinkedIn or Open ID - or open a Livefyre account using your email address.















 navigate econintersect.com

Blogs

Analysis Blog
News Blog
Investing Blog
Opinion Blog
Precious Metals Blog
Markets Blog
Video of the Day
Weather

Newspapers

Asia / Pacific
Europe
Middle East / Africa
Americas
USA Government
     

RSS Feeds / Social Media

Combined Econintersect Feed
Google+
Facebook
Twitter
Digg

Free Newsletter

Marketplace - Books & More

Economic Forecast

Content Contribution

Contact

About

  Top Economics Site

Investing.com Contributor TalkMarkets Contributor Finance Blogs Free PageRank Checker Active Search Results Google+

This Web Page by Steven Hansen ---- Copyright 2010 - 2016 Econintersect LLC - all rights reserved