Fernandez Rodrik (1991) - Resistance To Reform Status Quo Bias In The Presence Of Individual Specific Uncertainty

From edegan.com
Jump to navigation Jump to search
Article
Has bibtex key
Has article title Resistance To Reform Status Quo Bias In The Presence Of Individual Specific Uncertainty
Has author Fernandez Rodrik
Has year 1991
In journal
In volume
In number
Has pages
Has publisher
© edegan.com, 2016


Reference(s)

Fernandez, R. and D. Rodrik (1991), Resistance to Reform: Status Quo Bias in the Presence of Individual-Specific Uncertainty, American Economic Review 81, 1146-1155. pdf


Abstract

Why do governments so often fail to adopt policies which economists consider to be efficiency-enhancing? Our answer to this question relies on uncertainty regarding the distribution of gains and losses from reform. We show that there is a bias towards the status quo (and hence against efficiency-enhancing reforms) whenever some of the individual gainers and losers from reform cannot be identified beforehand. There are reforms which, once adopted, will receive adequate political support but would have failed to carry the day ex ante. The argument does not rely on risk aversion, irrationality, or hysteresis due to sunk costs.


Summary

This is a general equilibrium model. There are winners and losers, but it is unclear who will be a winner ex ante. This uncertainty leads workers to vote against reforms that may ex post provide benefits to a majority, and for reforms that may ex post provide benefits to only a minority. In an extension that allows for reforms to be overturned, this second case is eliminated. The basic problem is that ex ante it is not clear, to an individual, that an individual will be in the benefit recieving majority, leading to a status quo bias.


Key Assumptions

The following assumptions are used:

  • The voting uses majority rule - though other social choice mechanisms could apparently be used instead
    • A mechanism that costless translates the intensity of preferences into outcomes, such as costless lobbying, would undo the result and all efficient reform would be passed.
    • Voting is done ex post - to allow for negotiation of outcomes that are will make a majority worse off
  • The players are risk neutral - though risk aversion would make the result that the status quo is preferred even stronger.
  • Both the majority rule and risk neutrality are common knowledge
  • There must be uncertainty regarding who will benefit for the mechanism to work
  • Ex ante workers are identical and atomistic


The Model

Summary

The paper studies the choice of workers to invest in changing sectors. Part of their costs of changing sectors is fixed and known. The other part is private and unknown: Ex-ante, workers only know the distribution of types but not their own location within the distribution. As such, they act as if their own cost is the mean of the distribution.

In these circumstances, situations may arise in which a majority of workers oppose legislation that would require them to change sectors (such as trade liberalization, the motivating example of this paper) and instead favor the status quo. However should the legislation pass anyway, the workers would find out where they stand in the distribution. Some portion of workers who changed would then realize that the new legislation is actually better for them than the old status quo and would therefore resist efforts to repeal the law and revert to the old status quo.

Other situations may arise in which a majority of workers supports a similar piece of legislation because of an attractive mean. However upon learning their personal costs, a fraction of the voters who changed industries realizes that they may have preferred the status quo.

Details

There is a two sector perfectly competitive economy, where the sectors produce goods [math]X\,[/math] and [math]Y\,[/math] respectively. Both sectors use one factor, labour [math]L\,[/math], and have constant returns to scale.

[math]X = \frac{L_x}{a_x}\,[/math]
[math]Y = \frac{L_y}{a_y}\,[/math]

where [math]a_j \gt 0, j \in \{x,y\}\,[/math]

The cost of relocating between sectors for labour is made up of [math]\theta\,[/math], a general cost, and [math]c_j\,[/math], a sector specific entry cost. [math]c_j\,[/math] is unknown and drawn from [math]f(c)\,[/math], which is known. The switching sequence is that workers must expend [math]\theta\,[/math] in order to decide whether to switch, and if they do they then incur [math]c_j\,[/math].


By backwards induction a worker who has expended [math]\theta\,[/math] will switch to industry [math]X\,[/math] (WLOG) iff:

[math]c_x \le \tilde{c} = \tilde{w}_x - \tilde{w}_y\,[/math]


where [math]\tilde{w}_j\,[/math] is the equilibrium wage in sector [math]j\,[/math] following a reform.


Then, as workers are ex-ante identical, a worker will decide to incur the fixed cost [math]\theta\,[/math] iff:

[math](\tilde{w}_x - \tilde{w}_y)F(\tilde{c}) - \int_{\underline{c}}^{\overline{c}} f(c) c dc -\theta \ge 0\,[/math]


The paper assumes that the country is small with respect to world markets, so that relative prices are fixed by world price ratios. The country has an initial tariff:

[math]P_0 = \frac{a_x}{a_y}\,[/math]


where [math]P = \frac{p_x}{p_y}\,[/math] is the tariff inclusive price ratio. Let [math]Y\,[/math] be the imported good, with a domestic price normalized to 1. Then decreases in the tariff will increase the relative price of good [math]X\,[/math]. The intuition is:

  • The tariff goes up: suppose because [math]a_x\,[/math] goes up while a_y stays constant.
  • Suppose labour stays constant in both sectors, then by the definition of [math]X\,[/math] and [math]Y\,[/math], less [math]X\,[/math] is produced
  • Less [math]X\,[/math], in a perfectly competitive market, implies [math]p_x\,[/math] goes up.


The labour market's initial distribution between sectors is [math]L_x^0\,[/math] and [math]L_y^0\,[/math] and is taken as given. Perfect competition in the labour market means that:

[math]w_j = \frac{p_j}{a_j}\,[/math]


Therefore the initial tariff is [math]w_x^0 = w_y^0\,[/math]. Noting that [math]w_y = \frac{1}{a_y}\,[/math] and is invariant in [math]P\,[/math].


As the tariff falls:

  • The relative price of good [math]X\,[/math] goes down, and the demand goes up, requiring more labour. Therefore the wage in sector [math]X\,[/math] goes up.
  • [math]\tilde{w}_x - \tilde{w}_y\,[/math] increases
  • To start with no individual will choose to undertake the general investment cost
  • The value of [math]\tilde{c}\,[/math] increases as [math]\frac{d \tilde{c}}{dP} = \frac{d \tilde{w}_x}{dP} = \frac{1}{a_x}\,[/math]
  • The decision to undertake [math]\theta\,[/math] (above) is increasing in [math]P\,[/math]
  • At a sufficiently high [math]P^*\,[/math] all [math]Y\,[/math] sector individuals are indifferent between investing in [math]\theta\,[/math]
  • Those with a [math]c_i \le c*\,[/math] will move
  • Further increases will lead to further relocation


The paper wants circumstances where:

  • The reform would be accepted under certainty
  • But rejected under uncertainty (despite risk neutrality)
  • So suppose that the tariff is reduced such that the price ratio increases from [math]P^0 to P^*\,[/math]

This is:

[math]\tilde{c}F(\tilde{c}) - \int_{\underline{c}}^{\overline{c}} f(c) c dc -\theta = 0\,[/math]


If [math]L_x^0 \le L_y^0\,[/math] this measure would be rejected by a majority vote as:

  • Sector [math]Y\,[/math] individuals are exactly indifferent between trying to switch and not
  • Sector [math]Y\,[/math] individuals who remain in sector [math]Y\,[/math] have strictly less purchasing power in terms of good [math]X\,[/math] (in terms of good [math]Y\,[/math], their purchasing power is unchanged).

On the other hand, if the sector [math]Y\,[/math] individuals knew their identity under the new regime (i.e. if they knew their [math]c_i\,[/math] and were asked if they would be willing to switch at a cost of [math]\theta + c_i\,[/math]) then some would know that they would in sector [math]X\,[/math] and would be willing to vote for the regime change.


That is, it can be shown that there must exist some [math]c_i\,[/math] such that:

[math]v(P^*,w_x^* - \theta -c_i) \gt v(P^0,w_y^0)\,[/math]


In order to provide a clear example the paper assumes that preferences are given by:

[math]V(P,I) - v(P)I = \frac{I}{P^{\gamma}}\,[/math]


where [math]I\,[/math] is income and [math]1 \ge \gamma \gt 0\,[/math].


Furthermore [math]c\,[/math] is assumed [math]c \sim U[0,\overline(c)]\,[/math] such that [math]f(c) = \frac{1}{\overline{c}}\,[/math] and [math]\tilde{c} = (2 \theta \overline{c})^{\frac{1}{2}}\,[/math].


In this case it is noted that:

[math]w_x^* = \frac{P^*}{a_x} = w_y^* + \tilde{c} = w_y^0 + \tilde{c} = \frac{1}{a_y} + \tilde{c}\,[/math]
[math]\therefore P^* = P^0 + \tilde{c} a_x\,[/math]


The paper then shows that there exists a [math]c_i\,[/math] that satifies the condition above, and assuming the functional form above it notes that there are parameter values for which this is true (e.g. [math]a_y = \theta = 1, \tilde{c} = 2, \gamma = 0.5\,[/math]).

Dynamic considerations

Suppose now that there are two stages of voting, so that a reform that is a 'mistake' can be reversed by a majority vote. In the first stage the reform is voted on, in the second stage the reform can be continued or reversed.

The paper shows that:

  • Reforms instituted with majority support may be short lived
  • There is a tendancy towards inertia (i.e. the status quo)

There are four possibilities:

  • Reform is implemented and then reversed
  • Reform is implemented and sustained
  • Reform is always opposed
  • Reform is first rejected then accepted - this equilibrium is not possible unless the second period is lengthier than the first, and would require strategic voting (i.e. everyone voting against in the first, and for in the second, in order to preserve the form for a longer period). With equal periods, no information is revealed and there is no incentive to accept the reform in the second period. This equilibrium is hence forth ignored.

The paper then proceeds to provide an example of the first equilibrium above (i.e. implemented and then reversed). The intuition is as follows: At the start of the first period the worker make decisions much like before, but with a [math]P' \gt P^* \gt P^0\,[/math] to account for the increased wage earnings across two periods with discounting, and the calculation of [math]\tilde{c}\,[/math] also across two periods with discounting. Then if the following condition is met the reform will be reversed in the second period:

[math]F(\tilde{c})L_y^0 + L_x^0 \lt (1-F(\tilde{c}))L_y^0\,[/math]


The paper then checks that workers will rationally vote for the reform in the first period knowing that it will be overturned in the second period. This results in a more restrictive condition for voting for the reform in the first place (as now only one period's worth of temporary benefits can be obtained), but this is nevertheless achievable. The lower the discount factor the more likely it is that a reform that is accepted when permanent will still be accepted when temporary.