Welfare programmes, effective marginal tax rates, and the US marriage penalty

This week, my ECONS102 class covered the economics of social security. In countries like New Zealand and the US, the tax and transfer system (of which social security is one part) leads to a complicated relationship between income before taxes and transfers, and income after taxes and transfers are accounted for. The effective marginal tax rate (EMTR) is the amount of the next dollar of income a taxpayer earns that would be lost to taxation, decreases in rebates or subsidies, and decreases in government transfers (such as benefits, allowances, pensions, etc.) or other entitlements. Because of the variety of government programmes, benefits, and entitlements, it is not trivial to try to work out the effective marginal tax rate, and it varies widely based on circumstances. However, we should recognise that any time the EMTR exceeds 100 percent (or even when it is much lower than that), there will be significant incentive effects.

One example comes from this post by Ed Dolan on the Institute for Family Studies blog from earlier this year. Dolan notes the existence of a significant marriage penalty in the US:

What would happen if the two adults represented in Figure 1 moved in together to form a single cohabiting household with two adults and two children? What would happen if the cohabiting adults then married? 

Figure 2 shows the answer. It assumes that the two adults live in the home previously occupied by the parent with children, pool their incomes, and share all expenses. This time, the vertical lines are based on an FPL [Federal Poverty Line] of $27,750 for a four-person household. For convenience, the figure assumes that each of the two adults has approximately the same earnings...

The relation of net household resources to employment income differs dramatically for the two household configurations. Beginning from zero, the married couple at first does better. Total household resources are higher over most of the range up to the FPL. The married couple’s work incentives are also stronger. Over the range from zero to 100% of the FPL, net household resources rise by $1.28 cents for each dollar earned compared with $1.13 for the cohabiting couple. These advantages come partly from the fact that the EMTR and CTC phase in faster for the married couple, and partly because SNAP and health benefits do not phase out as quickly.

Beyond the FPL, however, the situation is reversed. Between earnings of $28,000 and $56,000, the red curve flattens dramatically as the married couple’s EMTR rises to a confiscatory 88%, compared to just 30% for the cohabiting couple. That is because SNAP, the EITC, and health benefits phase out simultaneously over this income range for the married couple. For the cohabiting couple, the phase-outs are spread over a much wider income range and overlap less. Due to the higher EMTR, net household resources for the married couple drop below those for the cohabiting pair soon after reaching the FPL. 

After earnings rise past twice the FPL, the difference in EMTRs essentially disappears, but the household resource gap never closes. Even when earnings reach $80,000 per year, the married couple is still worse off by more than $10,000.

The Figure 2 that Dolan refers to is shown below. The vertical axis shows net household resources (after taxes and transfers are accounted for), and the horizontal axis shows employment income. The dotted 45-degree line represents points where net household resources are equal to employment income (any transfers received from the government exactly offset taxes). When the solid lines are above the 45-degree line, the household receives more in transfers than they pay in taxes, and when the solid lines are below the 45-degree line, the household pays more in taxes than they receive in transfers. The EMTR is demonstrated by the slope of the lines (a higher EMTR is represented by a flatter slope). The marriage penalty is demonstrated by the fact that the red solid line is below the blue solid line beyond employment income of about US$30,000. This is mostly caused by the high EMTR for married couples from employment income of US$30,000 to US$55,000 (after that the slopes of the two lines are roughly the same).

It is clear from the figure that there is a substantial marriage penalty in the US, arising from how the broader system of taxes and transfers works. Dolan concludes that:

In short, although the welfare system gives a small marriage bonus to couples who are in deep poverty, it imposes a large marriage penalty on households that are just past the official poverty line but still striving to reach full self-sufficiency.

Fortunately, in New Zealand I don't think there is such a marriage penalty. Married and cohabitating parents are treated similarly, in terms of their entitlements. However, there has in the past been a penalty associated with cohabitating. For example, this 2019 report by Olivia Healey and Jennifer Curtin (both University of Auckland) notes that:

The weekly Supported Living Payment for those who are single with children is $379.19; compared to $237.09 each for those who are married, in a civil union or in a de facto relationship. Therefore, those who are partnered may be better off financially if they separated given two singles would receive a combined amount of $758.38 compared to $474.18 as a couple. The financial penalty on couples is a difference of $282.20 a week.

All countries with more than the simplest of all social security systems are likely to have a myriad of similar problems, which seem to raise issues of fairness and equity. The custodians of the social security system need to better identify the circumstances in which they will occur, consider what unintended consequences may arise, and then address those in a sensible way (which may include changing the rules to avoid these situations arising in the first place). Unless the goal really is to penalise married couples.