The household composition dimension of inequality and the role of welfare

There are different dimensions of economic inequality, and they require different types of policies to address them.

The three major sources of economic inequality:

1. Household composition.
2. Wages
3. Ownership.

Too often the different dimensions are smooshed together and this results in muddled thinking about inequality and sub-optimal policy. This post will focus on household composition inequality as it is the most overlooked, and most often muddled in public discourse, even among the “serious people”.

Household composition inequality is driven by the variation in the ratio of non-workers to workers by household. In an important sense it is prior to the other dimensions of inequality, which emerge in the market. We create household composition inequality literally by being born.

Yet inequality based on household composition is often overlooked. This is true in much popular discourse as well as the great treatises of political economy, with the grand clashes around factor payments – capital and market labour – drawing the attention. But roughly half the population are neither wage earners nor big time capitalists. These groups include old people, children, people with disabilities, carers, students and the unemployed. In discussions about inequality – even among socialists and other leftists – it sometimes seems like these people are invisible (feminist work is important exception).

To survive, these non-workers must rely on some combination of (1) personal savings, (2) private insurance, (3) intra-household transfers (support from workers in their household), or (4) welfare payments.

Personal savings and private insurance can help at times, but they are not systematic, sustainable and generalisable solutions for non-workers. They are also inefficient, as people don’t know much to save, and insurance markets in deeply human contexts like this struggle with information-related problems like adverse selection. Some financial risks are impossible to cover with an actuarial approach to insurance. Costs associated with events typically considered desirable by the bearer of the risk, such as child birth, are an example.

Intra-household transfers are probably the norm within many families – and they are standard between parents and young children – but they drag down the living standards of the supporting workers, failing to solve the underlying household composition inequality. A worker who lives alone may earn the same wage as another worker who supports four children and a disabled spouse, but the former household is far better off than the latter in per capita living standards.

Moreover, we cannot assume that intra-household transfers are equitable or benign. They may not always occur, or not occur in sufficient amounts, and it places the household breadwinner is a position of power over the dependents. This has the effect of reinforcing patriarchal domination, and parental domination over older children and young adults.

Luckily modern humans have invented an ingenious technology to solve all these problems – welfare payments. We don’t enough pause to consider just how remarkable and world changing an innovation this has been. Not only is it the one technology that can eliminate absolute and relative poverty in developed countries, it also serves wider egalitarian goals and provides valuable financial services.

Welfare pools the costs of non-workers, up to some minimum or egalitarian standard, across all of society’s workers and capitalists, particularly those on high incomes. This ensures households are funded for their non-workers, with the costs shared equitably across the population.

Welfare payments are made directly to the individual non-worker. If the recipient is not considered to have the mental ability to manage their own finances, it is paid to a guardian on their behalf. Usually this is the primary carer for a child, but a similar situation can occur with highly disabled and old people. Paying directly to the non-worker or their guardian will often help decentralise financial power within the household.

Only welfare payments can deliver a systematic and sustainable solution to the problem of non-workers and household composition inequality. These payments should not be means tested against the income or assets of other members of the household. From the perspective of economic inequality, it is always inferior to concentrate the costs of dependents on their household’s workers, for any income level, compared to pooling it across workers of a similar or higher income through progressive increases to taxation and universal welfare. In addition, it is the only way of providing people with any kind of income guarantee, unless legislators mandated workers transfer fixed sums to non-workers in their household. (This rather unlikely proposition would allow advocates to defensibly say they guarantee minimum incomes but would not address the underlying household composition inequality.) The universal approach is also the only way of consistently addressing the intra-household power dynamics discussed earlier.

Superficially, the universalistic model seems to create inefficiencies due to higher taxation. But this argument – sometimes referred to as “churn” – is not generally correct. The means testing approach simply replaces official taxes with the de facto taxes of income or assets tests, which are typically less efficient than broad-based progressive income taxation. As I wrote in my last post:

This concept can easily be illustrated in relation to child benefits (such as the Family Tax Benefit A in Australia). FTB A is withdrawn from middle- to upper-middle income parents who benefit, imposing an implicit tax. Similar regimes apply in most English speaking countries. An alternative approach would be to withdraw the same amount via an explicit tax progressively applied to all people above a certain level of income. An explicit tax can take in a far broader broader base for two reasons:

* a positive marginal tax rate can continue up to the top of the income distribution, whereas this would be impossible under means testing due to the inherently capped quality of benefit withdrawal.

*the taxable group is broadened from potential beneficiary parents to include all well off people, including non-parents.

By spreading the amount to be withdrawn over more people, the withdrawal rate – and thus the effective marginal tax rate – can be lowered. This more efficient approach – known as universalism – is how child benefits are administered in much of the developed world, and historically in the UK and Australia.

A largely overlapping way of looking at the universalistic welfare state is that it provides a type of financial service to individuals, which is to smooth out incomes over the life cycle as our ability to work changes. Given that most of us prefer income stability, this is a genuine gain to efficiency. Attempting to self fund for life’s exigencies is either impossible (childhood or youth ) or inefficient (we don’t know how much to save because we don’t know what will happen to us or how long we will live).

Welfare is a technology that is uniquely effective at addressing the household composition dimension of inequality. Other methods of changing distribution; such as full employment, unions or socialisation of ownership; are separately valuable but they address different dimensions of economic inequality, not household composition, as they do not target non-workers. Once a welfare payment is established, means testing leads to consistently inferior outcomes, both in terms of equity and efficiency, compared to the combination of welfare universalism and higher progressive taxation.