Problem 10 - Advanced Macroeconomics

Question

Uncertainty and policy. (Brainard, \(1967 .\) ) Suppose output is given by \(y=\) \(x+\left(k+\varepsilon_{k}\right) z+u,\) where \(z\) is some policy instrument controlled by the government and \(k\) is the expected value of the multiplier for that instrument. \(\varepsilon_{k}\) and \(u\) are independent, mean-zero disturbances that are unknown when the policymaker chooses \(z,\) and that have variances \(\sigma_{k}^{2}\) and \(\sigma_{u}^{2} .\) Finally, \(x\) is a disturbance that is known when \(z\) is chosen. The policymaker wants to \(\operatorname{minimize} E\left[\left(y-y^{*}\right)^{2}\right]\) (a) Find \(E\left[\left(y-y^{*}\right)^{2}\right]\) as a function of \(x, k, y^{*}, \sigma_{k}^{2},\) and \(\sigma_{u}^{2}\) (b) Find the first-order condition for \(z\), and solve for \(z\). \((c)\) How, if at all, does \(\sigma_{u}^{2}\) affect how policy should respond to shocks (that is, to the realized value of \(x\) )? Thus, how does uncertainty about the state of the economy affect the case for "fine-tuning"? (d) How, if at all, does \(\sigma_{k}^{2}\) affect how policy should respond to shocks (that is, to the realized value of \(x\) )? Thus, how does uncertainty about the effects of policy affect the case for "fine-tuning"?

Step-by-Step Solution

Answer

Uncertainty in \(u\) does not affect policy response, but higher \(\sigma_k^2\) reduces sensitivity to shocks.

Step 1: Express the Objective Function

The policymaker wants to minimize the expected square deviation: \[ E[(y - y^*)^2] \]. Substitute the expression for \(y\), which is \( y = x + (k + \varepsilon_k)z + u \), into the objective: \[ E[(x + (k + \varepsilon_k)z + u - y^*)^2] \].

Step 2: Expand the Objective Function

Expand the squared terms in the expression:\[ E[(x + kz + \varepsilon_k z + u - y^*)^2] = E[(x + kz + \varepsilon_k z + u - y^*)^2] \].This expands to:\[ E[(x + kz - y^* + \varepsilon_k z + u)^2] \].

Step 3: Simplify the Objective Function

Since \(\varepsilon_k\) and \(u\) are independent and have zero mean, the mean values can be factored out in expectation:\[ (x + kz - y^*)^2 + 2(x + kz - y^*)(\varepsilon_k z + u) + (\varepsilon_k z + u)^2 \].The expectation of cross-products involving \(\varepsilon_k\) and \(u\) are zero:\[ E[(x + kz - y^*)^2 + \varepsilon_k^2 z^2 + u^2] \].

Step 4: Apply Expectation and Variances

Apply the properties of expectations using variance:\[ E[(x + kz - y^*)^2] + E[\varepsilon_k^2] z^2 + E[u^2] \].Recall that \( E[\varepsilon_k^2] = \sigma_k^2 \) and \( E[u^2] = \sigma_u^2 \), so:\[ (x + kz - y^*)^2 + \sigma_k^2 z^2 + \sigma_u^2 \].

Step 5: Find the First-Order Condition

Take the derivative of the expanded form with respect to \(z\) to find the first-order condition:\[ \frac{d}{dz}[(x + kz - y^*)^2 + \sigma_k^2 z^2 + \sigma_u^2] \].Set the derivative to zero for minimization:\[ 2(x + kz - y^*)k + 2\sigma_k^2 z = 0 \].

Step 6: Solve for z

From the first-order condition, solve for \(z\):\[ 2k(x + kz - y^*) + 2\sigma_k^2 z = 0 \].Rearranging gives:\[ k(x - y^*) + (k^2 + \sigma_k^2)z = 0 \].Solving for \(z\) gives:\[ z = \frac{k(y^* - x)}{k^2 + \sigma_k^2} \].

Step 7: Analyze Influence of \(\sigma_u^2\)

Notice that \(\sigma_u^2\) does not appear in the expression for \(z\). This means that uncertainty in \(u\) does not affect the policy response to shocks. The policymaker cannot fine-tune the policy regarding the variance \(\sigma_u^2\).

Step 8: Analyze Influence of \(\sigma_k^2\)

The term \(\sigma_k^2\) appears in the denominator of the solution for \(z\). This implies that higher uncertainty about the policy multiplier (\(\sigma_k^2\)) decreases the sensitivity of the policy instrument \(z\) to shocks \(x\). Thus, greater uncertainty about policy effects discourages fine-tuning.

Key Concepts

Policy Multiplier

In economics, the policy multiplier is a critical concept that measures the effect of a policy instrument on output. In simple terms, it's akin to a lever that helps determine how much an economic policy can influence overall economic output. Specifically, in the context we're examining, the multiplier captures how changes in the policy instrument, like government spending or tax rates denoted as \( z \), translate into changes in output \( y \). The expected value of this interaction is represented by \( k \).

This effect isn't always certain; stochastic elements can cause the actual multiplier to deviate from expectations. These are represented by \( \varepsilon_k \), a random disturbance with a zero mean. Essentially, this reflects the unpredictability in how effective or ineffective a policy can be due to unforeseen factors. While economists can develop models to estimate the multiplier's effect, the real-world application often involves dealing with unexpected variances.

Stochastic Disturbances

Stochastic disturbances refer to random and unpredictable elements within an economic model that can affect outcomes. They are described by variables such as \( \varepsilon_k \) and \( u \) in our scenario. These disturbances have important implications for policy-making because they introduce uncertainty into the economic environment.

For example, \( \varepsilon_k \) represents the uncertainty in the policy multiplier, while \( u \) might represent other unforeseen shocks affecting the economy. What makes these disturbances particularly challenging is that their exact values are unknown at the time policy decisions are made. However, policymakers know their statistical properties, such as having a mean of zero and certain variances, \( \sigma_k^2 \) and \( \sigma_u^2 \), respectively.

This understanding allows policymakers to anticipate a range of possible outcomes, yet it complicates the process of fine-tuning policies as they must account for these unknown and potentially disruptive elements.

Fine-tuning in Macroeconomics

Fine-tuning in macroeconomics involves adjusting policy instruments to achieve desired economic outcomes, like full employment or stable prices. However, the presence of uncertainty complicates this process significantly. Fine-tuning requires precision and a clear understanding of how policy changes will ripple through the economy.

In our economic model, if the variance \( \sigma_u^2 \), representing the unpredictability of certain economic shocks, does not influence the solution for the policy variable \( z \), it suggests that these shocks are not amenable to fine-tuning. Thus, policymakers may find it ineffective to attempt adjustments based solely on such disturbances.

Conversely, the presence of \( \sigma_k^2 \) in determining the policy response indicates that as uncertainty about the multiplier increases, the ability to fine-tune becomes more constrained. Greater uncertainty makes it less effective to finely adjust policies, as the economic response to these changes is less predictable.

First-order Condition in Economics

The first-order condition in economics is a critical concept used in optimization problems to find the best level of a policy instrument. In the exercise, this involves determining the optimal amount of \( z \) so as to minimize the expected deviation of output \( y \) from its target \( y^* \).

This is done by taking the derivative of the objective function with respect to \( z \), setting this derivative to zero, and solving for \( z \). Mathematically, it looks like \( \frac{d}{dz}[(x + kz - y^*)^2 + \sigma_k^2 z^2 + \sigma_u^2] = 0 \).

The solution to this first-order condition gives insights into how responsive the policy instrument \( z \) is to economic shocks \( x \). Since \( \sigma_u^2 \) does not appear in the final expression for \( z \), it implies that the uncertainty introduced by \( u \) does not need adjustments in \( z \). On the other hand, \( \sigma_k^2 \) being in the denominator indicates that any increase in uncertainty about the multiplier like \( k \) will dampen \( z \)'s responsiveness. This mathematical tool is essential for understanding the practical implications of economic policies in uncertain environments.