John Irons's Blog


Economic News, Data and Analysis

Evaluating NAFTA

This past week marked the three year anniversary of the hotly debated
North America Free Trade Agreement (NAFTA). 
The debate over the agreement began anew with the release of the White
House’s review of the economic impacts of the NAFTA, and the finding that
the pact has had a “modest positive effect” on the economy. The debate
has changed from whether or not the NAFTA will be harmful to whether
or not is was harmful over the past three years. What is missing
from the debate is the following question: By what criteria should we judge
the success of the new policy?

The two sides of the debate have remained unchanged as have most of
the arguments. The proponents claim that NAFTA will create jobs and help
the economy via free trade, while opponents claim that jobs will be lost
to low wage workers in Mexico. The new report seems to have left the primary
debate substantially unaltered – proponents take it as evidence that the
agreement is good for the economy (since 311,000 trade-related jobs associated
with Mexico and Canada were created); and opponents site a swing from a
trade surplus to a trade deficit and a loss in worker bargaining power
as evidence that the economy (or a large subset of it) has been hurt.

There seems to be evidence supporting both sides of the debate. So how
are we to evaluate the net effect of NAFTA. In general, how should we evaluate
the economic effect of a major policy shift? The first necessary step is
to move away from the simple idea that the NAFTA is either “good” or “bad”.
As with any other major policy change there will be those who benefit from
the policy and those that lose.

“Good for the Economy”

When we say that something is “Good for the economy” what are we really
taking about? The net number of Jobs? The gross number of
jobs lost? or gained? The change in economic efficiency/productivity? The
average wage of Americans? The median wage of Mexicans? The total change
in GDP? The change in consumer prices?

Evaluation would be simple if the agreement were able to help (or harm)
everyone. But this is not the case – some benefit from the agreement (e.g.
shareholder and employees of companies that find new consumers in mexico,
and consumers that can purchase cheaper products) and some lose (employees
of companies that move plants to Mexico). Unfortunately, in this case the
economists favorite concept of a Pareto superior policy outcome (a policy
that helps at least one person with out hurting anyone else) does not apply.

Economists and Philosophers have invented a number of criterion with
which to evaluate alternative policies. For example we use the the Utilitarian
or Benthamite criterion where we would “add-up” or average the welfare
of the population and pick the policy which maximizes this sum. Another
possibility is the Rawlsian, or “maxi-min”, objective of maximizing the
welfare of the worst off individuals in the society. Another criteria would
be the equalization of welfare across every member of society, the so called
egalitarian outcome. This objective then runs into the problem of determining
what should in practice be equalized – incomes? opportunity?

Choosing from the criteria above is like asking who would win in a fight
– Batman or Spiderman. The concepts cannot be purely applied due to the
obvious problems of measuring and comparing individuals’ welfare; but we
can frame the debate in terms of who is helped or hurt by the policy rather
than whether or not the policy is “Good for the economy” in an abstract
and imprecise way.

For further reading on social objectives see the following.

Unknown Baseline

Another part of the problem we have in evaluating the effectiveness
of a policy change like NAFTA is that we are not running a controlled experiment.
It we had a economic laboratory we could take two identical North Americas
in 1994 – give one a NAFTA and leave the other one alone. We could then
compare the NAFTA-ized subject with its untouched counterpart to assess
the impact of the policy.

However, such experiments cannot be run and we are only left with one
subject to examine. The economic “baseline” of a NAFTA free North America
is unknown, we only have one set of numbers to use. So when we see some
statistic about, say, the number of jobs lost to NAFTA, we cannot be sure
if they were lost as a result of NAFTA, or whether they would have disappeared
anyway (say to China instead).

The economy is constantly in a state of flux, and it is very tricky
to isolate the impact of a single policy change independent of all the
other stuff going on. Since we do not know precisely what would have happened
with out the policy, even the “facts” of the NAFTA’s policy effects are
not clear cut.

Debate Continues

However the facts are eventually interpreted, it would be beneficial
to the debate to have a clear idea the criteria by which we will judge
the policy to be a success. Right now we have the two sides of the debate
using different welfare criteria to judge the policy – so it should be
no surprise that there is continued debate over the success of the policy.

See Also:

Filed under: International

Some Simple Economics of the Tobacco Deal

Why the price of cigarettes will not rise.


What effect will the recent tobacco deal have on the price of tobacco
products? There seems to be a popular misconception that the large fine
will force tobacco producers to raise prices to “recoup their losses.”
In other words, the settlement is of little consequence since the tobacco
companies can simply raise their prices and raise the extra cash from their
addicted consumers. But this logic doesn’t quite hold water. Barring changes
in supply factors, the price of tobacco products should actually fall.

There are basically two components of the deal. The first is a $368
billion payment by the tobacco companies to avoid future punitive damages.
The second component includes myriad changes in the regulation of tobacco
and tobacco advertising.

Neither of these components spells an increase in the price of tobacco;
and the second set of provisions are likely (since that is the goal) to
reduce demand for tobacco related products, and thus are likely to reduce

1) Lump sums don’t matter.

Let’s consider first the effect of the payments that the tobacco companies
must make as art of the deal. It is probably an open question whether or
not the net payment made by the tobacco companies over time is greater
than zero – they are settling a large number of potential lawsuits, which
saves them money via litigation costs as well as the amount of potential
jury awards. However, we’ll assume for the moment, that they will have
to pay a significant net amount out of their pockets.

The important aspect of this payment is that it is a lump sum.
In other words, the amount that the tobacco companies must pay does not
depend upon any actions that the company can take. In particular, it does
not depend upon the quantity of output that the firms produce. Lump sum
payments, in general, do not affect the optimal behavior of firms.

To illustrate, consider a typical tobacco company (which) is maximizing
profits, has some market power, and faces a fixed cost per unit of output).
Before the payment, the company sets the price on their cigarette brand
to maximize profits. Figure 1 shows the profits of the company as a function
of the price. Initially profits rise as the company increases the price
since the revenue received per unit sold is increasing faster than the
loss in customers (or due to decreasing returns to scale). Eventually,
as the price gets higher, the revenue received by the company begins to
decrease as customers are lost due to competition from other companies
and due to smokers quitting. In Figures 1, the maximum amount of revenue,
R*, is obtained at a price of P*.

Figure 2 shows the profits of the company after the payment of a lump
sum (of the amount x). The profit hill just shifts vertically downward
by the amount of the lump sum payment. As the graph demonstrates, the price
that maximizes the profits of the company, P*, remains unchanged.
Another way of saying this is to ask, if it is possible for the tobacco
companies to raise a portion of the additional $368 billion by raising
their prices, why have they not done so already to boost profits?

2) Decrease in demand

The second major component of the deal involves increasing the regulation
of tobacco products and decreasing the visibility of tobacco advertising.
The policies include altering the warning labels on boxes, funding anti-smoking
programs, and setting guidelines on advertising style (e.g. no more cartoons
or humans in ads). Most of the components of the deal are aimed at reducing
the demand for tobacco products.

To the extent that the regulatory burden is a lump sum, the argument
from above applies and the effect on the price should be minimal. Thus,
if the demand reduction policies work, we can expect the price to decrease
as the tobacco companies struggle in the face of weakening demand to retain


There are, of course, several caveats to the above
arguments. The most important of which arises from the peculiar market
structure of the industry. The above arguments for why the price will fall
assume that the market acts like a competitive market or a monopoly. If,
on the other hand, the market is characterized by a significant degree
of collusion (but not perfect collusion, in which case the industry would
operate like a monopolist), then the deal may have other effects.

In particular, if the negotiation process and the resulting settlement
were to somehow strengthen the degree of collusion – either explicit or
implicit – among the tobacco companies, then we would perhaps see an increase
in the prices of their products. So if there is a dramatic rise in the
price of tobacco in response to the settlement, the tobacco companies may
have to deal with the Justice department as well as the FDA.

See Also:



Other Caveats: 

Lump Sum 

  • Collusion: The lump sum payments may serve as a focal point for increasing
    prices if the tobacco companies are implicitly colluding to set prices. 

  • Liquidity constraints: If the tobacco companies cannot borrow freely in
    the market, then a reduction in their cash reserves may affect other parts
    of their operations (such as R&D). 

  • Business failure: The above assumes that no tobacco businesses fail as
    a result of the deal. Failures of some of the companies could lead to a
    change towards a less competitive industry.

Demand Reduction 

  • Composition of demand: Those that are likely to quit smoking first as overall
    demand drops first may have different price response characteristics (e.g.
    they may be more sensitive to price changes since they are “less addicted”).
    So as the pool of potential customers changes, so too will the composition
    of those in the market. The price elasticity of demand for those in the
    market may potentially fall; leading to a potential price increase. 

  • Intertemporal demand changes: If the future demand for tobacco dies up
    (e.g. if the pool of potential new smokers dries up), then tobacco companies
    will have little incentive to keep prices low in order to woo new smokers.
    Their incentive is then to fully exploit the current addicts since their
    future market is already gone. 



Filed under: Microeconomics