28/10/17: Income Inequality: Millennials vs Baby Boomers

OECD's recent report, "Preventing Ageing Unequally", has a wealth of data and analysis relating to old-age poverty and demographic dynamics in terms of poverty evolution. One striking chart from the report shows changes in income inequality across two key demographic cohorts: the Baby Boomers (born at the start of the second half of the 20th Century) and the Millennials (born in the last two decades of the 20th Century):

Source: http://www.oecd.org/employment/preventing-ageing-unequally-9789264279087-en.htm.

The differences between two generations, controlling for age, are striking. In my opinion, the dramatic increase in income inequality across two generations in the majority of OECD economies (caveats to Ireland and Greece dynamics, and a major outliers of Switzerland, France and the Netherlands aside) is one of the core drivers for changing perceptions of the legitimacy of the democratic ethics and values when it comes to public perceptions of democracy. 

You can read more on the latter set of issues in our recent paper, here: http://trueeconomics.blogspot.com/2017/09/7917-millennials-support-for-liberal.html.

The dynamics of income inequality for the Millennials do not appear to relate to unemployment, but rather to the job markets outcomes (which seemingly are becoming more polarized between high quality jobs/careers and low quality ones):

In other words, where as in the 1950s it was sufficient to have a job to gain a place on a social progression ladder, today younger workers need to have the job (at Google, or Goldman Sachs, or other 'star' employers) to achieve the same.

Thus, as low unemployment swept across the advanced economies in the post-Global Financial Crisis recovery, there has not been a symmetric amelioration of the youth poverty rates in a number of countries:

In 25 OECD countries out of 35, poverty rates for those aged 18-25 are today higher than for those of age 65-75. Across the OECD, statistically, poverty rates for the 18-25 year olds cohort are on par with those for of 76+ year olds cohort, and both are above 12 percent. 

There is a lot that is still missing in the above comparatives. For example, the above numbers do not adjust for differences between different age groups in terms of quality of health and education. Younger workers are also healthier, as a cohort, than older population groups. This means that their incomes should be expected to be higher than older workers, simply by virtue of better health.  Younger workers are also better educated than their older counterparts, especially if we consider the same age cohorts for current Millennials and the Baby Boomers. Which also implies that their incomes should be higher and their income inequality should be lower than that for the Baby Boomers.

In other words, simple comparatives under-estimate the extent of income inequality and poverty incidence and depth for the Millennials by excluding adjustments for health and education differences.

3/10/17: Ambiguity Fun: Perceptions of Rationality?

Here is a very insightful and worth studying set of plots showing the perceived range of probabilities under subjective measure scenarios. Source: https://github.com/zonination/perceptions

The charts above speak volumes about both, our (human) behavioural biases in assessing probabilities of events and the nature of subjective distributions.

First on the former. As our students (in all of my courses, from Introductory Statistics, to Business Economics, to advanced courses of Behavioural Finance and Economics, Investment Analysis and Risk & Resilience) would have learned (to a varying degree of insight and complexity), the world of Rational expectations relies (amongst other assumptions) on the assumption that we, as decision-makers, are capable of perfectly assessing true probabilities of uncertain outcomes. And as we all have learned in these classes, we are not capable of doing this, in part due to informational asymmetries, in part due to behavioural biases and so on. 

The charts above clearly show this. There is a general trend in people assigning increasingly lower probabilities to less likely events, and increasingly larger probabilities to more likely ones. So far, good news for rationality. The range (spread) of assignments also becomes narrower as we move to the tails (lower and higher probabilities assigned), so the degree of confidence in assessment increases. Which is also good news for rationality. 

But at that, evidence of rationality falls. 

Firstly, note the S-shaped nature of distributions from higher assigned probabilities to lower. Clearly, our perceptions of probability are non-linear, with decline in the rate of likelihoods assignments being steeper in the middle of perceptions of probabilities than in the extremes. This is inconsistent with rationality, which implies linear trend. 

Secondly, there is a notable kick-back in the Assigned Probability distribution for Highly Unlikely and Chances Are Slight types of perceptions. This can be due to ambiguity in wording of these perceptions (order can be viewed differently, with Highly Unlikely being precedent to Almost No Chance ordering and Chances Are Slight being precedent to Highly Unlikely. Still, there is a lot of oscillations in other ordering pairs (e.g. Unlikely —> Probably Not —> Little Chance; and We Believe —> Probably. This also consistent with ambiguity - which is a violation of rationality.

Thirdly, not a single distribution of assigned probabilities by perception follows a bell-shaped ‘normal’ curve. Not for a single category of perceptions. All distributions are skewed, almost all have extreme value ‘bubbles’, majority have multiple local modes etc. This is yet another piece of evidence against rational expectations.

There are severe outliers in all perceptions categories. Some (e.g. in the case of ‘Probably Not’ category appear to be largely due to errors that can be induced by ambiguous ranking of the category or due to judgement errors. Others, e.g. in the case of “We Doubt” category appear to be systemic and influential. Dispersion of assignments seems to be following the ambiguity pattern, with higher ambiguity (tails) categories inducing greater dispersion. But, interestingly, there also appears to be stronger ambiguity in the lower range of perceptions (from “We Doubt” to “Highly Unlikely”) than in the upper range. This can be ‘natural’ or ‘rational’ if we think that less likely event signifier is more ambiguous. But the same holds for more likely events too (see range from “We Believe” to “Likely” and “Highly Likely”).

There are many more points worth discussing in the context of this exercise. But on the net, the data suggests that the rational expectations view of our ability to assess true probabilities of uncertain outcomes is faulty not only at the level of the tail events that are patently identifiable as ‘unlikely’, but also in the range of tail events that should be ‘nearly certain’. In other words, ambiguity is tangible in our decision making. 

Note: it is also worth noting that the above evidence suggests that we tend to treat inversely certainty (tails) and uncertainty (centre of perceptions and assignment choices) to what can be expected under rational expectations:

In rational setting, perceptions that carry indeterminate outruns should have greater dispersion of values for assigned probabilities: if something is is "almost evenly" distributed, it should be harder for us to form a consistent judgement as to how probable such an outrun can be. Especially compared to something that is either "highly unlikely" (aka, quite certain not to occur) and something that is "highly likely" (aka, quite certain to occur). The data above suggests the opposite.

28/7/17: Risk, Uncertainty and Markets

I have warned about the asymmetric relationship between markets volatility and leverage inherent in lower volatility targeting strategies, such as risk-parity, CTAs, etc for some years now, including in 2015 posting for GoldCore (here: http://www.goldcore.com/us/gold-blog/goldcore-quarterly-review-by-dr-constantin-gurdgiev/). And recently, JPMorgan research came out with a more dire warning:

This is apt and timely, especially because volatility (implied - VIX, realized - actual bi-directional or semi-var based) and uncertainty (implied metrics and tail events frequencies) have been traveling in the opposite direction  for some time.

Which means (1) increasing (trend) uncertainty is coinciding with decreasing implied risks perceptions in the markets.

Meanwhile, markets indices are co-trending with uncertainty:

Which means (2) increasing markets valuations are underpricing uncertainty, while focusing on decreasing risk perceptions.

In other words, both barrels of the proverbial gun are now loaded, when it comes to anyone exposed to leverage.

7/6/17: European Policy Uncertainty: Still Above Pre-Crisis Averages

As noted in the previous post, covering the topic of continued mis-pricing by equity markets of policy uncertainties, much of the decline in the Global Economic Policy Uncertainty Index has been accounted for by a drop in European countries’ EPUIs. Here are some details:

In May 2017, EPU indices for France, Germany, Spain and the UK have dropped significantly, primarily on the news relating to French elections and the moderation in Brexit discussions (displaced, temporarily, by the domestic election). Further moderation was probably due to elevated level of news traffic relating to President Trump’s NATO visit. Italy’s index rose marginally.

Overall, European Index was down at 161.6 at the end of May, showing a significant drop from April 252.9 reading and down on cycle high of 393.0 recorded in November 2016. The index is now well below longer-term cycle trend line (chart below). 

However, latest drop is confirming overall extreme degree of uncertainty volatility over the last 18 months, and thus remains insufficient to reverse the upward trend in the ‘fourth’ regime period (chart below).

Despite post-election moderation, France continues to lead EPUI to the upside, while Germany and Italy remain two drivers of policy uncertainty moderation. This is confirmed by the period averages chart below:

Overall, levels of European policy uncertainty remain well-above pre-2009 averages, even following the latest index moderation.

7/6/17: Equity Markets Continue to Mis-price Policy Risks

There has been some moderation in the overall levels of Economic Policy Uncertainty, globally, over the course of May. The decline was primarily driven by European Uncertainty index falling toward longer-term average (see later post) and brings overall Global EPU Index in line with longer term trend (upward sloping):

This meant that short-term correlation between VIX and Global EPUI remained in positive territory for the second month in a row, breaking negative correlations trend established from October 2015 on.

The trends in underlying volatility of both VIS and Global EPUI remained largely the same:

The key to the above data is that equity markets risk perceptions remain divorced from political risks and uncertainties reflected in the Global EPUI. This is even more apparent when we consider actual equity indices as done below:

Both, on longer-run trend comparative and on shorter term level analysis bases, both S&P 500 and NASDAQ Composite react in the exactly opposite direction to Global Economic Policy Uncertainty measure: rising uncertainty in the longer run is correlated with rising equities valuations.

28/4/17: The VUCA Markets

My regular column with the Cayman Financial Review covering the current development of the risk-uncertainty frameworks in the markets is now available here: http://www.caymanfinancialreview.com/2017/04/26/welcome-to-the-vuca-world/.

28/2/17: Sentix Euro Breakup Contagion Risk Index Explodes

Sentix Euro Break-up Contagion Index - a market measure of the contagion risk from one or more countries leaving the euro area within the next 12 months period - has hit its post-2012 record recently, reaching 47.6 marker, up on 25 trough in 2Q 2016:

Key drivers: Greece, Italy and France.

Details here: https://www.sentix.de/index.php/sentix-Euro-Break-up-Index-News/euro-break-up-index-die-gefaehrlichen-drei.html.

23/2/17: Welcome to the VUCA World

Much has been said recently about the collapse of ‘risk gauges’ in the financial markets, especially on foot of the historically low readings for the markets’ ‘fear index’, VIX. In terms of medium-term averages, current VIX readings are closely matching the readings for the period of ‘peak’ ‘Great Moderation’ of 1Q 2005 - 4Q 2006, while on-trend, VIX is currently running below 2005-2006 troughs. In other words, risk has effectively disappeared from the investors’ (or rather traders and active managers) radars (see chart below).

At the same time, traditional perceptions of risk in the financial markets have been replaced by a sky-rocketing uncertainty surrounding the real economy, and especially, economic policies. The Economic Policy Uncertainty Indices have been hitting all-time highs globally (see chart below), and across a range of key economies (see this for my recent analysis for Europe: http://trueeconomics.blogspot.com/2017/01/15117-2016-was-year-of-records-breaking.html, this for Russia and the U.S.: http://trueeconomics.blogspot.com/2017/01/17117-russian-economic-policy.html). In current data, Economic Policy Uncertainty Index (EPUI) has been showing extreme volatility coupled with extreme valuations. Index values are rising above historical norms both in terms of medium-term averages and in terms of longer term trends.

 Another interesting feature is the direct relationship between the EPUI and VIX indices. Based on rolling correlations analysis (see chart below), the traditionally positive correlation between the two indices has broken down around the start of 2Q 2016 and since then all three measures of correlation - the 6-months, the 12-months and the 24-months rolling correlations - have trended to the downside, turning negative with the start of 2H 2016. Since November 2016, we have a four months period when all three correlations are in the negative territory, the first time this happened since June 2007 and only the second time this happened in history of both series (since January 1997). Worse, the previous episode of all three correlations being negative lasted only two months (June and July 2007), while the current episode is already 4 months long.

Final point worth making is that while volatility of VIX has collapsed both on trend and in level terms since the start of H1 2016 (see chart below), volatility in EPUI has shot up to historical highs.

Taken together, the three empirical observations identified above suggest that the current markets and economies are no longer consistent with increased traditional risk environment (environment of measurable and manageable risks), but instead represent VUCA (volatile, uncertain, complex and ambiguous) environment. The VUCA environment, by its nature, is characterised by low predictability of risks, with uncertainty and ambiguity driving down efficacy of traditional models for risk assessments and making less valid traditional tools for risk management. Things are getting increasingly more complex and uncertain, unpredictable and unmanageable.

16/9/2014: Mapping Uncertainty Across Industries

A very interesting post on HBR Blog (http://blogs.hbr.org/2014/09/the-industries-plagued-by-the-most-uncertainty/) mapping technological uncertainty against demand uncertainty across major industries.

Two charts:

11/9/2012: Inherent limit to artificial intelligence?

In a rather common departure from economics (as defined by rational expectations subset of the discipline) on this blog - here's a fascinating thinking about the artificial intelligence and the bounds of model-induced systems.

Especially close to me, as it explores that which I thought about back in 2003-2004 when I wrote an essay on the role of leaps of faith (irrational and discontinuous jumps in human creativity and thinking) as the foundation for humanity and, thus, a foundation for recognition of the property rights over uncertainty.

Economics 16/10/2010: Organizational systems and uncertainty

I came across this very interesting, and to me - far reaching - paper on the effects of organizational structures on the organization's ability to cope with uncertainty and change. Karynne L. Turner, Mona V. Makhija. “Measuring what you know: an individual information processing perspective” (April 15, 2010). Atlanta Competitive Advantage Conference 2010 Paper (here).

According to the information processing perspective, the organization’s ability to draw upon and utilize information is dependent on the relationship between structure and the ability of individuals to process information, facilitated by specific organizational aspects of the firm. The study considers the effect of two types of structure, organic (integrated or systemic) and mechanistic (siloed), on individuals’ ability to gather, interpret and synthesize information, and their problem-solving orientation. Evidence shows that individuals develop more information processing capability under organic than mechanistic structures, which in turn creates more problem solving orientation in individuals.

In short, the study lends support to the premise that better integrated, more diversified across skills and less siloed organizations produce more effective and efficient gathering, processing and interpreting of information, as well as better problem solving.


Effective management of knowledge is the basis of firms’ ability to compete (Zander and Kogut, 1995; Nonaka, 1994). This is achieved through organizational design (Teece at al., 1997) that underlies “the means by which firms acquire, disseminate, interpret and integrate organizational knowledge”.

Organizational structure embodies a number of key elements, such as control and coordination or management mechanisms, and human capital management that allocate tasks to work units and individuals, and coordinate them in a way that achieves organizational goals. The manner in which this is done is critical due to problems created by

• External uncertainty associated with suppliers, competitors and consumer demand (Gresov and Drazin, 1997; Sine, Mitsuhashi and Kirsch, 2006), or
• Internal uncertainty, due to the complexity of internal coordination, measurement difficulties and changing processes (Habib and Victor, 1991).
  

Uncertainty reduces the effectiveness of pre-established routines, technologies or goals, and increases the importance of problem solving (Becker and Baloff, 1969)). The more work related uncertainty increases, the greater the need there will be for information processing (Turner and Makhija, 2006 and Tushman, 1979).

One way in which an organization addresses uncertainty is by assigning specific responsibilities to specialized subunits, which collect, process and distribute information acting as “a set of nested systems” (Daft and Weick, 1984).

Literature distinguishes two types of organizational structures, mechanistic and organic. These structures differ in the distribution of tasks, the flow of information among individuals and across units, and the extent to which there is interaction with the environment (Shremata, 2000; Gibson and Birkinshaw, 2004).

Mechanistic forms of organization are characterized by hierarchical division of labor, in which communication tends to be in one direction – top to bottom. Individuals develop deep expertise in their own designated jobs, which tend to be clearly specified and specialized in individual knowledge. The mechanistic structures do not allow for much flexibility (Parthasarthy and Sethi, 1993).

Organic forms of organizations are based on horizontally-administered teams, in which all members participate in management decisions (Baum and Wally, 2003), allowing for worker autonomy, responsibilities adaptation. Team members developing competence across multiple tasks, thus diversifying their skills and knowledge sets. Individuals have broader unit-level knowledge rather than just one job and develop greater flexibility.

The structural differences between mechanistic and organic organizational forms are likely to influence the development of information processing capability in organizational members, reflected in organization’s ability to gather, interpret and synthesize information. Turner and Makhija (2010) consider the impact of different types of structures on each of these three aspects of organizational members’ information processing capability.

Turner and Makhija (2010) postulate a set of testable hypotheses all of which are confirmed:

H1: Organic structures lead to more gathering of information than mechanistic structures.
Implication: uncertainty is reduced in organic (integrated or more horizontal) structures through reduced information asymmetries vis-à-vis external environment.

H2: Organic structures lead to more similarly interpreted information than mechanistic structures.
Implication: information asymmetries are reduced across the broader range of the organization structures in the organic setting.

H3: Organic structures lead to more synthesized information than mechanistic structures.
Implication: organic systems are better capable of integrating information of various types.

H4: More gathering of information is associated with greater problem solving orientation.
Implication: organic systems are better able to cope with converting uncertainty into manageable risks systems.

H5: More similarly interpreted information is associated with greater problem solving orientation.
Implication: better information processing in organic systems results in better problem solving, so information is used more effectively.

H6: More synthesized knowledge is associated with greater problem solving orientation.
Implication: individuals also tended to synthesize, or understand the interrelationships among different types of information, much better than individuals working in mechanistic structures.

H7: Information processing capability mediates the relationship between organizational design and problem solving orientation
Implication: the effects of individuals’ information processing on their problem solving orientation is greater in the organic structures, reflecting their comfort with problem situations in their work.

Turner and Makhija (2010) research shows that, when operating in two different types of structures, individuals process information differently in all three respects: gathering, interpreting, and synthesizing information.

These findings have several far-reaching implications for the organizational structures found in Ireland.

Firstly, it is clear that hierarchical and fixed systems approach to public services provision – characterized by the lack of communications between vertically-integrated public sector departments and organizations leads to their inherently lower ability to absorb, process and implement informational processes that manage uncertainty.

Secondly, this shows why successful entrepreneurial ventures are horizontal in nature and less siloed.

Third, it shows that our political system – with disproportionate powers allocated to the executive, as opposed to more uniform distribution of powers between the executive, legislative and judiciary – is similarly to the public sector less equipped to handle uncertainty.