Can Fiscal Stimulus Work?
Last year, the federal government passed a massive stimulus bill, spending approximately $787 billion. The bill was highly controversial, with several of the best economists in the world on either side of the debate (Pro: Krugman, Bernanke, Summers, DeLong); Against: Lucas, Buchanan, Becker, Mankiw). Any one of those economists is world class and highly intelligent, although they disagree on such a fundamental issue. At first glance, it's like if half the Nobel prize winning physicists signed a petition saying the rocket engines on NASA shuttles should face the other way on the day before the launch. On closer inspection, it's more like the debate over string theory - there's no hard evidence either way. We are left to use intuition and a few historical examples to see what might be true. For the purposes of this essay, I define "fiscal stimulus" as debt financed spending. Seigniorage (printing money) financed spending is too tied up in monetary policy, and should be addressed in a separate essay.
There are two sides to every macroeconomic story - the nominal and the real. The nominal (demand) side concerns money - where does it come from, how is it spent, how fast does it move and who pays the bills. When thinking about the nominal side of macro, it is helpful to think about Pigou wealth effects, balance sheets, and inflation. Demand side economics is the study of how to get the overall amount of spending to the correct level. The real (supply) side of macroeconomics has to do with production - how much and where people work, how resources are allocated, how capital is allocated and what regulations affect the economy. Supply side economics is the study of how to get the composition of what is made correct and to maximize production. Both sides are important to understanding the economy as a whole.
Recessions from the demand side:
The purpose of any stimulus is to increase the rate of spending - more people spending more money. In the short run, more spending leads to more output. People work overtime, companies produce more to feed the increased demand. Another reason for this is because prices are sticky; they take some time to change. Some goods' prices change rapidly, like stocks and commodities. Other goods' prices change only slowly, like debt and wages. If the prices are stuck, then quantities must adjust instead. In the long run, more spending leads to inflation. Production will eventually hit the limits of scarcity and firms will bid up the prices of inputs, like raw materials and capital. The higher prices then increase the cost of production and the price of final goods.
Reductions in the rate of spending work essentially the same, but in reverse. Firms confuse a general reduction in spending for a reduction in spending for their sector in particular and scale back production. For example, if spending drops 10%, a firm making cars will see the reduced sales and think that customers don't like their cars as much. Workers will be laid off and capital will sit idle. Debt is also highly impacted by changes in the rate of spending. Firms' incomes drop, so it becomes harder for them to pay back loans they took out to purchase capital during the boom. If they cannot repay their debt, they may be forced into bankruptcy, which is extremely expensive. Banks in particular are very susceptible to nominal shocks (a fancy economist word for increases and decreases in the rate of spending). They typically have tons of debt and have long term investments. Some banks were leveraged 30 to 1 during the housing boom (see chart below), which means if they took a 3.34% loss to their investments, they would become bankrupt. Wages are also sensitive to decreases in spending, because workers don't like it when their wages are cut. Morale suffers when wages are cut, even if deflation occurs simultaneously. Because of price stickiness, changes in the rate of spending have real effects on the economy.
Recessions from the supply side:
There are two main schools of thought for the supply side of recessions - Real Business Cycle Theory (RBCT) and Austrian Business Cycle Theory. Each of them deserve their own post, but in a nutshell, Real BCT emphasizes productivity shocks , and Austrian BCT emphasizes how changes in the interest rate can affect the profitability of capital investments. In a capitalist economy, sectors are continually growing a shrinking. Recessions frequently involve shifts from one industry to another at a higher rate than normal.
Source: Vernon Smith
Arguments in favor of Fiscal Stimulus:
Nominal side:
During fiscal stimulus, the government raises money through the sale of bonds and then spends it. If it is able to spend the money faster than the purchasers of the bonds would have, and other spending stays the same or speeds up, total spending will increase. Those who buy bonds willingly forgo buying goods now. By buying a bond, a person demonstrates that they would rather have money in the future than money today. Thus, it is likely that purchasers of bonds would not have spent the money they lent to the government. Issuing bonds may also increase spending above and beyond the amount spent, because bonds are similar to currency in many ways. A bank can use them as collateral for a loan or use them to purchase financial assets. If those who buy the bonds use them the same way as they would money, fiscal stimulus spending will not be offset by a decline in private spending.
If people suffer a large loss to their stock of wealth, they will try to rebuild it. In normal times, they can save more. During a recession, if they all decide to save more in currency (and not invest in capital), they will lower consumption, but not increase capital by a corresponding amount. Eventually prices will drop to increase the real value of money, but in the short term, output will drop. Krugman wrote a good story explaining how this can happen. Note that in his example, many factors that alleviate the business cycle in the real world were absent - price adjustments, monetary stimulus and capital formation.
Real Side:
During recessions, there are unemployed resources - not just workers, but capital as well. Advocates of fiscal stimulus say that the government can target the unemployed resources and use them productively. Thus, output would expand without affecting the employed part of the economy and would not generate any inflation. Also, if prices are sticky, output and employment will change before prices do, and the economy will experience a recovery from any increased spending (either monetary or fiscal).
Against:
Nominal Side:
- People expect higher taxation when the bonds mature, so they may save more money to survive in the future high tax environment (since they won't be able to earn as much).
- The purchasers of bonds might have otherwise spent the money either on consumption or capital formation. If they would have invested the money in capital, it could have increased the productive capacity of the economy.
- If the monetary authority is targeting inflation, they can offset fiscal stimulus. When Congress adds more to the government budget, the Fed can take out as much money as it wants simultaneously. Congress can do nothing if the Fed wants to inflate or deflate, since they do not command the level of resources the Fed does.
- When Congress wants to spend more money, they must design a bill, debate on it, and pass it. Then they must start the money rolling through the bureaucracy. It can be years before the money is finally spent - by which time the recession might already be over. As of September 2009, 20% of the stimulus had been spent - and the recession was already in effect for over a year and a half. Even today, only 60% has been spent, and the recession is officially over.
Real Side:
The market is a discovery process. Companies do not know what people want - they must experiment and figure it out. When the recession is due to a sectoral shift - more of one thing and less of another, the government does not know how far the new trend will go. For example, after the housing bubble collapse, most of the unemployment was limited (at first) to the construction sector. If the government tries to save the old jobs, they will simply prolong the adjustment process.
Fiscal stimulus, unlike private expenditure, might not conform to the wants of the people. Since speed is highly important in a business cycle, projects are often not well thought out and conform to political pressure rather than maximizing returns or productivity. While fiscal stimulus might increase measured GDP, it might not make very many people subjectively better off.
If the government conscripted half of the US population to dig holes all day and conscripted the other half to fill them back in, and paid each of us a billion dollars a day for the task, and valued holes that were dug and holes that were filled in at a trillion dollars a hole, then GDP would be very very large, unemployment would be zero and there would be no stimulating effect and we would soon be dead from starvation.
http://johnbtaylorsblog.blogspot.com/2010/12/stimulus-math-many-multiples-of-nothing.html
There are two sides to every macroeconomic story - the nominal and the real. The nominal (demand) side concerns money - where does it come from, how is it spent, how fast does it move and who pays the bills. When thinking about the nominal side of macro, it is helpful to think about Pigou wealth effects, balance sheets, and inflation. Demand side economics is the study of how to get the overall amount of spending to the correct level. The real (supply) side of macroeconomics has to do with production - how much and where people work, how resources are allocated, how capital is allocated and what regulations affect the economy. Supply side economics is the study of how to get the composition of what is made correct and to maximize production. Both sides are important to understanding the economy as a whole.
Recessions from the demand side:
The purpose of any stimulus is to increase the rate of spending - more people spending more money. In the short run, more spending leads to more output. People work overtime, companies produce more to feed the increased demand. Another reason for this is because prices are sticky; they take some time to change. Some goods' prices change rapidly, like stocks and commodities. Other goods' prices change only slowly, like debt and wages. If the prices are stuck, then quantities must adjust instead. In the long run, more spending leads to inflation. Production will eventually hit the limits of scarcity and firms will bid up the prices of inputs, like raw materials and capital. The higher prices then increase the cost of production and the price of final goods.
Reductions in the rate of spending work essentially the same, but in reverse. Firms confuse a general reduction in spending for a reduction in spending for their sector in particular and scale back production. For example, if spending drops 10%, a firm making cars will see the reduced sales and think that customers don't like their cars as much. Workers will be laid off and capital will sit idle. Debt is also highly impacted by changes in the rate of spending. Firms' incomes drop, so it becomes harder for them to pay back loans they took out to purchase capital during the boom. If they cannot repay their debt, they may be forced into bankruptcy, which is extremely expensive. Banks in particular are very susceptible to nominal shocks (a fancy economist word for increases and decreases in the rate of spending). They typically have tons of debt and have long term investments. Some banks were leveraged 30 to 1 during the housing boom (see chart below), which means if they took a 3.34% loss to their investments, they would become bankrupt. Wages are also sensitive to decreases in spending, because workers don't like it when their wages are cut. Morale suffers when wages are cut, even if deflation occurs simultaneously. Because of price stickiness, changes in the rate of spending have real effects on the economy.
Recessions from the supply side:
There are two main schools of thought for the supply side of recessions - Real Business Cycle Theory (RBCT) and Austrian Business Cycle Theory. Each of them deserve their own post, but in a nutshell, Real BCT emphasizes productivity shocks , and Austrian BCT emphasizes how changes in the interest rate can affect the profitability of capital investments. In a capitalist economy, sectors are continually growing a shrinking. Recessions frequently involve shifts from one industry to another at a higher rate than normal.
Source: Vernon Smith
Arguments in favor of Fiscal Stimulus:
Nominal side:
During fiscal stimulus, the government raises money through the sale of bonds and then spends it. If it is able to spend the money faster than the purchasers of the bonds would have, and other spending stays the same or speeds up, total spending will increase. Those who buy bonds willingly forgo buying goods now. By buying a bond, a person demonstrates that they would rather have money in the future than money today. Thus, it is likely that purchasers of bonds would not have spent the money they lent to the government. Issuing bonds may also increase spending above and beyond the amount spent, because bonds are similar to currency in many ways. A bank can use them as collateral for a loan or use them to purchase financial assets. If those who buy the bonds use them the same way as they would money, fiscal stimulus spending will not be offset by a decline in private spending.
If people suffer a large loss to their stock of wealth, they will try to rebuild it. In normal times, they can save more. During a recession, if they all decide to save more in currency (and not invest in capital), they will lower consumption, but not increase capital by a corresponding amount. Eventually prices will drop to increase the real value of money, but in the short term, output will drop. Krugman wrote a good story explaining how this can happen. Note that in his example, many factors that alleviate the business cycle in the real world were absent - price adjustments, monetary stimulus and capital formation.
Real Side:
During recessions, there are unemployed resources - not just workers, but capital as well. Advocates of fiscal stimulus say that the government can target the unemployed resources and use them productively. Thus, output would expand without affecting the employed part of the economy and would not generate any inflation. Also, if prices are sticky, output and employment will change before prices do, and the economy will experience a recovery from any increased spending (either monetary or fiscal).
Against:
Nominal Side:
- People expect higher taxation when the bonds mature, so they may save more money to survive in the future high tax environment (since they won't be able to earn as much).
- The purchasers of bonds might have otherwise spent the money either on consumption or capital formation. If they would have invested the money in capital, it could have increased the productive capacity of the economy.
- If the monetary authority is targeting inflation, they can offset fiscal stimulus. When Congress adds more to the government budget, the Fed can take out as much money as it wants simultaneously. Congress can do nothing if the Fed wants to inflate or deflate, since they do not command the level of resources the Fed does.
- When Congress wants to spend more money, they must design a bill, debate on it, and pass it. Then they must start the money rolling through the bureaucracy. It can be years before the money is finally spent - by which time the recession might already be over. As of September 2009, 20% of the stimulus had been spent - and the recession was already in effect for over a year and a half. Even today, only 60% has been spent, and the recession is officially over.
Real Side:
The market is a discovery process. Companies do not know what people want - they must experiment and figure it out. When the recession is due to a sectoral shift - more of one thing and less of another, the government does not know how far the new trend will go. For example, after the housing bubble collapse, most of the unemployment was limited (at first) to the construction sector. If the government tries to save the old jobs, they will simply prolong the adjustment process.
Fiscal stimulus, unlike private expenditure, might not conform to the wants of the people. Since speed is highly important in a business cycle, projects are often not well thought out and conform to political pressure rather than maximizing returns or productivity. While fiscal stimulus might increase measured GDP, it might not make very many people subjectively better off.
If the government conscripted half of the US population to dig holes all day and conscripted the other half to fill them back in, and paid each of us a billion dollars a day for the task, and valued holes that were dug and holes that were filled in at a trillion dollars a hole, then GDP would be very very large, unemployment would be zero and there would be no stimulating effect and we would soon be dead from starvation.
http://johnbtaylorsblog.blogspot.com/2010/12/stimulus-math-many-multiples-of-nothing.html