2. Evidence and theories on the drivers of inequality

http://www.guardian.co.uk/business/2008/jan/17/economics.economy
Stock markets have more impact on the inequality in Britain than the government's tax and benefits policies, a thinktank report says today.

http://www.ifs.org.uk/publications.php?publication_id=4107

focuses on income inequality and the incomes of the super rich, boosted by the stock market.

Bernard (2008)

"The VOC [Varieties of Capitalism] literature suggests that the institutional arrangements that make up natinal production regimes - financial arrangements and corporate governance, industrial relations, education and training systems, and welfare state - and the institutional complementarities that these institutions generate when operating together produce important micro- and macro-economic effects (Hall and Soskice 2001)." -

But though " financial arrangements and corporate governance" are first on the list, this is the last time in the whole paper that they're mentioned!

"The limitation of the VOC perspective in accounting for cross-national variance in inequality is that its explanatory leverage is observable primary on income (i.e., wage) distribution prior to tax and transfers." - p1. So it doesn't tell us much about wealth inequality, income through ownership and redistribution through tax and benefits.

"Figure 1 uses data from Statistics Canada to track the evolution of income inequality and redistributive profile of taxes and benefits in Canada between 1976 and 2004." p4 - find sim data for UK money flows analysis.

p6 - LMEs are characterised by arms length, short term, flexible relationships.
"SMEs are characterized by longer term, trust and obligation-based relationships. Such arrangements bring parties (employers and employees, banks and firms, management and shareholders, government, and producer groups) closer together, creating an environment where such parties have a stake in each others' activities... and maintain a long-term horizon."

Large body of VOC thinkers suggest that "divergent patterns of distributional equality among advanced industrial countries are owed in large part to the.. wage bargaining system, educational and training system, and the social protection." p6

"Earnings dispersion, by far the most important determinant of the overall distribution of income, is closely related to particular skill systems as well as the wag ebargaining institutions that tend to go with these systmes. By themselves this part of dichotomous variables account for nearly 70 per cent of the cross-national variance in income inequality." (Esteves-Abe et al (2001: 177, 181 and 182)

basic summary of the observation / arg:

"In SMEs, where htere is high quality and redistribution, wage setting institutions are centralised and coordinated, acquisition of asset-specific skills in the workforce is prevalent, and the welfare state provides generous social protection. In the low equality, low redistribution LMEs, wage-setting institutions are decentralised and uncoordinated, general skills constitute the core sklls of the workforce, and the welfare state provides limited protection against economic risk." p6

(how much of an issue is wealth inequality from ownership? How do we place it among the issues / drivers?)

Union density has dramatically decreased in LMEs and only slightly decreased in SMEs between 1980 and 2000.
















SMEs have more vocational training than LMEs. This raises the marketable skills and earning potential of lower earning population segments. There is a "strong negative association between wage inequality and vocational training intensity." Scattergraph on page 9.

"Whereas a large proportion of early school leavers in [specific skills systems] acquire valuable skills through the vocational training system, in [general skill systems] most early school leavers end up as low-paid unskilled workers for most or all of their working lives." p9 (author notes that the logic of this argument has been criticised)

p10 - data suggests that higher public spending on education is associated with lower levels of income inequality.

Research by Burkhart (2005) concludes that "Capitalism has a positive effect on the income distribution variable, meaning that increasing levels of capitalism leads to more income controlled by the top 20 per cent of the population." P16 HOwever, he suggests further research to "discern what exactly it is about capitalism that makes it influential in models of democratic performance and income distribution." p19

Beinhocker (2007)

Epstein and Axtell (1996) "model that included nothing more than people with a few basic abilities, and an environment with some natural resources.

Virtual island with only one resource - sugar. Fifty by fifty grid. each square on the grid has different amounts of sugar on it, from 0 to 4. Two sugar mountains, with 'badlands' in between.

The Sugarscape agents could make decisions and take actions. Simple version: look for sugar, move, eat sugar. + a metabolism for digesting sugar. The agent is credited by the amount of sugar eaten and debited by the amount of sugar burned - so can build surplus. If sugar goes below zero, they've starved to death.

Agents are heterogenous - they differ by vision length and fast (bad) or slow (good) metabolisms.

The distribution of sugar among the agents began as a bell curve and ended in a Pareto curve, a long tail. the rich get richer effect. "The skewed distribution is an emergent property of the system." p86


Evidence on global wealth transfers from poor to rich countries:

• In 2001, according to nef's research, developing countries paid rich countries:
• $122bn in interest payments on loans
• $55bn in profits sent back home by multinationals
• (interest and profits alone represent about 3% of their GDP...)
  
• $260bn loan principle
• $110bn through the forced holding of foreign currency reserves, usually US dollars (such transfers 'constitute an immends and expanding transfer of wealth out of these economices' to the United States" - p36)
• $122.4bn "illegal and thus unrecorded capital flight from poor countries"...? nef (2003:37)

• "Trickle down is a discredited idea. Let us think instead of globalisation as a 'hoover' effect, sucking resources away from poor people in both poor and rich countries, and concentrating them in the pockets of the rich." nef (2003:37)

• This suggests that debt is a bigger drain of capital from poor to rich countries than trade. But debt is historically specific. It can be cancelled. It is not - need to double check this thinking but - it's not part of the system design. The wealth drain from trade is part of system design and will continue until the system design is improved.

Interest

From The Future of Money, Lietaer 2001 p53

• A German study investigated the transfer of wealth between income groups through interest payments. They grouped the population into ten groups of about 2.5m households each. As you can see from the graph, the bottom 80% had a net 'loss' of interest, and the top 20% had a net gain, which was huge for the top 10% and slight for the 9th decile.

• The top 10% of households recieved a net transfer of DM 34.2 billion in interest from the rest of the society during 1982. The greatest transfers occurred from the middle classes (deciles 3 - 8) which each transferred about DM 5 billion to the top 10%. Even the lowest interest households transferred DM 1.8 billion. (orig. source: Kennedy, Margrit: Interest and Inflation Free Money. Sava International 1995, p26)

• I would like to do something similar with mapping out the resource flows around the case study companies and their stakeholders

• Leitaer: "This graph clearly shows the systematic transfer of welath from the bottom 80% of the population to the top 10%. This transfer was due exclusively to the monetary system in use, and is completely independent of the degree of cleverness or industriousness of the participants - the classic argument to justify large differences in income." (p54)

Banks et al (IFS, 2000)

• Aim to understand differences in financial wealth between the US and the UK, and conclude thus: "potential candidates for investigation arising from this paper have been the stock market, annuity markets and the housing market, all of which could be important explanatory factors in generating measured wealth differences between the two countries." They explore and discount the major importance of income,
  
• strong evidence that stock ownership is a primary way for people to accrue wealth - eg Tables 5 and 7 in the annex, powerful data at individual but not aggregate level. "Throughout, increments in total financial net assets are almost one to one with teh magnitude of the capital gains achieved in the stock market." p24
  
• "there is little doubt the stock market surge was largely responsible for increasing wealth inequaltiy in the United States during the 1980s and 1990s." p24
  
• large inequalities in wealth in both countries, greater in US.
  
• In seeking explanation: consider income inequalities, which explain some but by no means all of the wealth inequality. They also explore and discount inheritance and measurement issues.
  
• "Only recently (post 1988) has the UK had substantial levels of direct share ownership in comparison to the US, and even then the direct holdings of equity of many stockholders are small." p6
• The data does not provide any measure of private pension or government pension wealth. Data sourced primarily from British Household Panel Study (BHPS) and hte Panel Study of Income Dynamics in the US, supplemented by others inc Financial Research Survey and Family Expenditure Survey in the UK and the Survey of Consumer Expenditures in the US. Externally to this study, the PSID contains data on business equity while the BHPS does not- observe implication that "both surveys understate the concentration of wealth among the extremely wealthy." p9 "Both surveys exclude the super-rich." "To the extent that omitted components vary across countries, and particularly for groups converting business wealth to personal wealth, these may be important issues which deserve further investigation." p9
• "We begin with two concepts of household wealth - total household wealth (excluding business equity) - and total financial assets." p11 Point of confusion: they do include wealth in shares and bonds in the total financial assets. Equity the same as shares, no? equity considered in financial assets then? details p10
  
• "Simple summary statistics such as means and medians can be quite misleading when the subject is wealth" because wealth can be distributed so unevenly among any given grouping
• "Total household wealth is about a third higher in the US" p11. "On average, in the mid 1990s American households owned about $20,000 more in corporate equity." p12
• wealth distribution, US and UK, figures in the annex. In USD rather than % of total, but could calculate. Organised by deciles and age groups - for 1995, 1984-2005. Lots of data on the distribution of share wealth, particularly in table 5, (1994 BHPS - shareholders in the 50th percentile had an average (what kind of average?) of US$10,100 held. Overall average in 50th percentile was US$0.0. Shareholders in the 95th percentile: US$156,800,000. 98th percentile - US$326,000,000 - excluding pensions wealth. p59. Could graph this data but only starts at 50th percentile, with average wealth in stock and shares at 0. Table 7 - 'changes in financial assets and capital gains in stocks' - period? - shows strong relationship - strong evidence that stock ownership is a primary way for people to accrue wealth. Shows how much wealth ends up there at the individual level, but not at the national level (if had data on how many people each percentile represents could calculate the latter from the former, but is aggregate data available elsewhere?)
• 'The UK index is the FInancial Times All Share index
• Figure 8 shows percentage of UK population owning shares, - from 9% in 1978 to 23% in 1996.
  
• (What was British gas's distributive pattern prior to privatisation?)
• p28 - considering reasons for differences in share ownership patterns US-UK. One explanation involves "differences in attitudes toward capitalist financial institutions. Especially during the 1970s and early 1980s, it is probably a fair characterisation that there was more distrust of the fairness of capitalism as an economic system at least among significant segments of the European population. The stock market is one of hte most vivid capitalist symbols..."p28
• Table 10 - could make a graph of it - share ownership and political differences in the UK. p64. Conservative voters hold a great deal more share wealth than labour voters.

where this leaves us:
- fairly strong evidence here that share ownership is a signficant driver of inequalities in wealth. Methodology- doesn't relate it to change, solution or policy options. So what do we do with that? Depends on political objectives - are we ok with inequality or not? Not related to company ownership model so scope to research that aspect. Data provided in terms of average wealth per household in specific categories, rather than stock and flows of wealth at the national or even global levels.

So, yet to seek data on:
the average income of different wealth percentiles
stocks and flows of wealth at the national level
relationship between the wealth distribution trends reported here and company ownership models

____

If models of business ownership had an impact on inequality then we might expect to see a correlation between a country's patterns of business ownership and it's wealth distribution.

I plotted the data for the capitalisation of listed companies as a % of GDP against data about the % of a nation's wealth held by the top 10% of the population, and from a visual look there doesn't seem to be a correlation. There are many other variables that may have an influence, including, for example, the extent to which investment tends to be global or national in scope.

http://www.guardian.co.uk/commentisfree/2007/dec/05/theweathofthenation

"Rising inequality can only be addressed by a major change in the political system and the way companies are governed"

The median annual pre-tax wage for a full-time adult worker is estimated (pdf) to be £23,764, a rise of 2.9 per cent since 2006. The bottom 10 per cent of the employees receives around £13,000 a year. At the same time, the remuneration of executives at major companies has doubled in the last five years. Though the Labour government has sought to manage poverty and inequality through tax, pension credits and a variety of mean-tested benefits, no political party has put forward any programme to tackle the root causes.

(which are, he says, our majority vote political system rather than the scandanavian consensus model which incentivises politicians to respond to minority demands)

In contrast to Scandinavia and even Germany and France, the UK corporate structures lack two-tier boards, works councils and effective trade unions. UK employees are not allowed to elect directors and have no say in how the wealth generated with their own blood and sweat should be distributed.

A study (pdf) by the Organisation for Economic Co-operation and Development stated: "a stronger bargaining power of trade unions is associated with lower relative poverty and income inequality."

(But his emphasis on inequality is still on wage inequality:)

In 1975, at the dawn of Thatcherism, 65.1 per cent of the gross domestic product (GDP) went to workers in the shape of salaries and wages. By 1996, it declined to 52.6 per cent though with the introduction of minimum wage and investment in public sector it rose to 55.6 per cent in 2006, still nearly 10 per cent less than in 1975. The shrinking share of the cake has been sliced unevenly with the fat cats taking the biggest share. The result is an inevitable increase in inequality.

Prem Sikka is Professor of Accounting at the University of Essex