In last week’s article we looked at the relation between disposable income and private consumption. The basic conclusion was that a higher disposable income allows higher levels of consumption, but we also noted that in general people tend to save a larger portion of their disposable income when they make good money.
Already here, we can point to a policy conclusion that is hardly controversial: since disposable income is the money we have left of our earnings after taxes, higher taxes will depress private consumption. This, in turn, has two consequences:
Today’s article explains the second point. It also concludes with tax-policy recommendations based on the analysis of state-level consumption data.
A key variable here is poverty, which can be measured in many different ways. The most often used measurement states that a person lives in poverty if he makes no more than 60 percent of median income. A person who sees his income decline can therefore be said to become “poorer” because his income is falling relative median income.
Another way to measure poverty is to look at the content of a person’s private consumption expenditures. The larger the share of our personal consumption expenditures that go toward essential items, the poorer we are.
Essential items are things that we need to buy just to continue a basic standard of living: food, shelter (think “apartment lease”) and utilities, motor vehicles and gasoline, clothes, furnishings and appliances, and health care. In a non-industrialized country that list would look differently, but in a nation at the height of industrialized living it is not surprising that the idea of essential living expenditures is a bit different.
There is a practical, fiscal policy meaning to the concept of “essential consumption”. The more of our money we spend on non-essentials, i.e., things that make our lives comfortable, even add a luxurious touch, the easier it is for us to reduce spending when times get tough. In technical econo-speak: the price and income elasticities of demand for non-essential items are high: we reduce spending quickly when prices go up or our income falls.
By contrast, we are far less inclined to reduce our consumption of essential items. When our disposable income declines, we may try to substitute Pasta Barilla for spaghetti from Sam’s Club, but it is far more difficult to cut away substantial costs of living out of essential items than non-essentials.
If government wants to increase taxes on persons by means of, e.g., higher personal income taxes, then it better have a good idea of the consumption pattern in their jurisdiction. In a state where essential consumption is relatively small, an increase in income taxes - reducing disposable income - will bring about a quick response in terms of reduced higher-end consumption. The effects on the upscale retail market can be significant, if the tax increase is big enough. Reduced spending can bring down sales-tax revenue much more quickly than income-tax hiking legislators realize.
In a state where consumption is predominantly oriented toward essentials, consumer response is much less pronounced - and the repercussions for government are different. When disposable income declines, households will do everything they can to keep spending unchanged. They may cut other things than essential living expenses, such as savings and retirement planning.
However, the most important consequence for the state government will be that people living with small margins are encouraged to go apply for various kinds of tax-paid assistance. Soon enough, the state legislators who increased personal income taxes may find themselves having to fund higher costs for low-income energy assistance, food stamps, housing assistance, etc.
But are there really any differences in our country as to how people spend their money? Sure, there are low-income and high-income families, but are they not spread evenly across all states?
No. There are surprisingly clear differences in consumption patterns across the country. Table 1 reports per-capita personal consumption expenditures (PCE) in current-price dollars for, respectively, 1997 and 2014 (the percentages are explained below the table):
Source: Bureau of Economic Analysis
The difference in PCE between the District of Columbia and Mississippi is an astounding 90 percent: the individual consumer in the District of Columbia spends $1.90 for every dollar that the individual consumer in Mississippi doles out.
Based on the earlier reasoning about essential consumption and comfort/luxury spending, it would make sense that Mississippians spend a larger share of their money on essentials than the consumers in the District of Columbia. To find out, let us look at the percentages in Table 1 and compare them to per-capita private consumption. The result for 1997 is reported in Chart 1:
Source: Bureau of Labor Statistics
The red function displays PCE per capita per state. The states are ranked along the horizontal line in the same order as they are in the section of Table 1 that represents 1997. The scale for the red line is displayed on the left vertical axis. The light blue line reports the share of PCE in each state that is spent on essentials; the blue line is a trend line for the essentials share.
Even though the essentials share varies a bit between the states independently of total PCE, there is a clear trend. Once a state’s per-capita consumption falls below $20,000 the share spent on essentials goes up rather quickly.
So far we have reviewed data from 1997. Over now to the same data from 2014 as shown in Chart 2:
Source: Bureau of Economic Analysis
Here, the negative correlation is more pronounced. The essentials share starts increasing earlier than it did in 1997 (though the amplitude - the individual variation between the states - is also stronger).
With two years that far apart we can conclude, in three steps:
1. There are major differences in per-capita consumption between the states.
2. The share of consumption expenditures allocated to essential items varies negatively with per-capita consumption.
3. State legislators are well advised to study private consumption patterns in their state.
The last point is crucial. As mentioned earlier, tax increases can have immediate, and tangible, repercussions for both the state economy and the state government budget. But so do tax cuts.
There is an important implication for tax reform in the data reported here: if state legislators want to reduce government spending, they can do so with relative ease if they first reduce personal taxes. By cutting those taxes they increase the disposable income of the residents of their state. As disposable income increases, consumers migrate in consumption upward from essentials to comfort and luxury spending. As they do, they signal reduced dependency on the state’s welfare programs.
This is a genuine opportunity to implement reforms to permanently reduce the cost of welfare-related entitlements: as more families can support themselves, fewer will depend on public or private handouts.
Notably, this “trade” between personal income taxes and welfare costs is less obvious in states with relatively prosperous consumers. For those states, our data suggest a strong opportunity for shifting the state’s tax burden from, e.g., personal income tax to sales taxes. In a wealthy state, a cut in personal income taxes will send more people into the upper layers of consumption, thus boosting sales tax revenue.
Sven R. Larson is an economist and Member of the Council of Scholars with Compact for America. He has a blog called America’s Fiscal Future and he recently published "Robbing the Millennials: How We Looted Our Kids' Future and the New Handshake We Owe Them"