What Is the Laffer Curve?
The Laffer Curve is a theory formalized by supply-side economist Arthur Laffer to show the relationship between tax rates and the amount of tax revenue collected by governments. The curve is used to illustrate the argument that sometimes cutting tax rates can result in increased total tax revenue.
Understanding the Laffer Curve
The Laffer Curve is based on the economic idea that people will adjust their behavior in the face of the incentives created by income tax rates. Higher-income tax rates decrease the incentive to work and invest compared to lower rates. If this effect is large enough, it means that at some tax rate, and further increase in the rate will actually lead to a decrease in total tax revenue. For every type of tax, there is a threshold rate above which the incentive to produce more diminishes, thereby reducing the amount of revenue the government receives.
At a 0% tax rate, tax revenue would obviously be zero. As tax rates increase from low levels, tax revenue collected by the also government increases. Eventually, if tax rates reached 100 percent, shown as the far right on the Laffer Curve, all people would choose not to work because everything they earned would go to the government.
It's thus necessarily true that at some point in the range where tax revenue is positive, it must reach a maximum point. This is represented by T* on the graph below. To the left of T*, an increase in tax rate raises more revenue than is lost to offsetting worker and investor behavior. Increasing rates beyond T*, however, would cause people not to work as much or not at all, thereby reducing total tax revenue.
Therefore, at any tax rate to the right of T*, a reduction in tax rate will actually increase total revenue. The shape of the Laffer Curve, and thus the location of T* is dependent on worker and investor preferences for work, leisure, and income, as well as technology and other economic factors.
Governments would like to be at point T* because it is the point at which the government collects the maximum amount of tax revenue while people continue to work hard. If the current tax rate is to the right of T*, then lowering the tax rate will both stimulate economic growth by increasing incentives to work and invest, and increase government revenue because more work and investment means a larger tax base.
The Laffer Curve Explained
The first presentation of the Laffer Curve was performed on a paper napkin back in 1974 when its author was speaking with senior staff members of President Gerald Ford’s administration about a proposed tax rate increase in the midst of a period of economic malaise that had engulfed the country. At the time, most believed that an increase in tax rates would increase tax revenue.
Laffer countered that the more money was taken from a business out of each additional dollar of income in the form of taxes, the less money it will be willing to invest. A business is more likely to find ways to protect its capital from taxation or to relocate all or a part of its operations overseas.
Investors are less likely to risk their capital if a larger percentage of their profits are taken. When workers see an increasing portion of their paychecks taken due to increased efforts on their part, they will lose the incentive to work harder. Put together these could all mean less total revenue coming in if tax rates were raised.
Laffer further argued that the economic effects of reducing incentives to work and invest by raising tax rates would be damaging in the best of times and even worse in the midst of a stagnant economy. This theory, supply-side economics, later became a cornerstone of President Ronald Reagan’s economic policy, which resulted in one of the biggest tax cuts in history. During his time in office, annual federal government current tax receipts from $344 billion in 1980 to $550 billion in 1988, and the economy boomed.
Is the Laffer Curve Too Simple a Theory?
There are some fundamental problems with the Laffer Curve—notably that it is far too simplistic in its assumptions. First, that the optimal tax revenue-maximizing tax rate T* is unique and static, or at least stable. Second that the shape of the Laffer Curve, at least in the vicinity of the current tax rate and T* is known or even knowable to policymakers. Lastly, that maximizing or even increasing tax revenue is a desirable policy goal.
In the first case, the existence and position of T* depend entirely on the shape of the Laffer Curve. The underlying concept of the Laffer Curve only requires that tax revenue be zero at 0% and at 100%, and positive in between. It says nothing about the specific shape of the curve at points in between 0% and 100% or the position of T*.
The shape of the actual Laffer Curve might be dramatically different from the simple, single peaked curve commonly depicted. If the curve has multiple peaks, flat spots, or discontinuities, then multiple T*’s might exist. If the curve is skewed deeply to the left or right, T* might occur at extreme tax rates like 1% tax rate or a 99% tax rate, which might put tax revenue-maximizing policy into serious conflict with social equity or other policy goals.
Furthermore, just as the basic concept does not necessarily imply a simply shaped curve, it does not imply that a Laffer Curve of any shape would be static. The Laffer Curve might easily shift and change shape over time, which would mean that to maximize revenue, or just avoid falling revenue, policymakers would have to constantly adjust tax rates.
This leads to the second criticism, that policymakers would be in practice unable to observe the shape of the Laffer Curve, the location of T*, whether multiple T*’s exist, or whether and how the Laffer Curve might shift over time. The only thing policymakers can reliably observe is the current tax rate and associated revenue receipts (and past combinations of rates and revenue).
Economists can guess what the shape might be, but only trial and error could actually reveal the true shape of the curve, and only at those tax rates that are actually implemented. Raising or lowering tax rates might move the rate toward T*, or it might not. Moreover, if the Laffer Curve has any shape other than the assumed simple, single peaked parabola, then tax revenue at points between the current tax rate and T* could have any range of values higher or lower than revenue at the current rate and the same or lower than T*.
An increase in tax revenue after a rate change would not necessarily signal that the new rate is closer to T* (nor a decrease in revenue signal that it is further away). Even worse, because tax policy changes are made and applied over time, the shape of the Laffer Curve could shift; policymakers could never know if an increase in tax revenue in response to a tax rate change represented a movement along the Laffer Curve toward T*, or a shift in the Laffer Curve itself, with a new T*. Policymakers trying to reach T* would effectively be groping in the dark after a moving target.
Lastly, it is not clear on economic grounds that maximizing or increasing government revenue (by moving toward T* on the Laffer Curve) is even an appropriate goal for choosing tax rates. It might easily be the case that a government could meet the otherwise unmet needs of its citizens and provide any necessary public goods at some level of revenue lower than the maximum it can potentially extract from the economy, perhaps much lower depending on the position of T*. If so, then given the well-researched principal-agent problems, rent-seeking, and knowledge problems that arise with the politically driven allocation of resources, putting additional funds in public coffers beyond this socially optimal level might just produce additional unnecessary social costs, inefficiencies, and dead-weight losses.
Maximizing government tax revenue by taxing at T* would also likely maximize these costs. A more appropriate goal might be to reach the minimum tax revenue necessary to achieve only those socially necessary policy goals, which would seem to be almost the exact opposite of the purpose of the Laffer Curve.
David James Connolly