Please Share:: 
India Equity Research Reports, IPO and Stock News
Visit http://indiaer.blogspot.com/ for complete details �� ��
Goldman Sachs Research
How Much Downside?
Exhibit 1 summarizes average outcomes during the eleven post-war US recessions. In a typical recession, real GDP contracts by 2.3% peak-to-trough. In the mildest US recession (2001), GDP fell by just 0.3% and actually rose over the year as a whole. In the most severe recession (2007-09) real GDP contracted by 5.1%. On average consumption has fallen by 0.6%, and nominal corporate profits by about 20%.
Recessions are also hard on the labor market. On average nonfarm payroll employment has declined by 2.8% peak-to-trough in the past. At the current level of employment, this would imply a loss of about 3.7 million jobs. In the mildest past recession for the labor market (1980), nonfarm payrolls fell by 0.9%—the equivalent of 1.2m jobs today. The unemployment rate typically rises by about 3 percentage points (pp) and has always risen by at least 1.9pp.
Any recession would clearly be harmful, but there have been a wide range of outcomes in the past. The severity of a recession would of course depend on the size of the shock hitting the economy—a large fiscal contraction would be worse than a small one, for example. But initial conditions probably matter as well. We think three main factors are important today: 1) the level of activity in cyclical parts of the economy, 2) the stance of monetary and fiscal policy, and 3) the state of the financial system.
Low Activity in Cyclical Sectors
The pullback in spending during recessions is typically concentrated in just a few segments of the economy: consumer spending on motor vehicles and other durable goods, residential investment, business fixed investment, and inventories. These highly cyclical sectors provide virtually all of the amplitude in GDP growth (Exhibits 2 & 3).
Because of the deep decline in output during the last recession and moderate recovery to date, activity in cyclical sectors remains relatively low. We argued last year that low levels of activity in the most cyclical parts of the economy reduce the odds of recession. In our view, recessions are partly corrections of imbalances that build up in the economy over time. If activity is still recovering from an earlier recession and those imbalances have not yet built up, growth may be more resilient to any given shock. Low activity in cyclical sectors is also likely to mitigate the severity of a potential recession: with spending still weak, it may be difficult to cut further in some areas.
The clearest example is housing. After declining by roughly 60% since its peak at the end of 2005, residential investment is only barely above the rate of depreciation in the housing capital stock (Exhibit 4). In other words, although housing investment has not fallen to zero, it is probably near its practical minimum as current investment is mostly just covering wear and tear. Despite ups and downs in the broader recovery, the main housing indicators—housing starts, new home sales, builder sentiment, etc.—have mostly remained flat at low levels, which also suggests the housing market has probably reached a floor. If the economy enters recession, we suspect that residential investment will not fall much further.
A similar argument likely holds for spending on motor vehicles. Vehicle sales are currently running at an annualized rate of just over 12 million units. This is close to the annual “scrappage” rate for vehicles in recent years, implying that recent sales merely keep the number of cars on the road constant. Lower sales are certainly possible, but from the current weak level this would require a smaller auto stock or a gradual deterioration in quality through depreciation.
Stress Test
To quantify the potential downside in cyclical components of final expenditure we estimated the impact on GDP if cyclical spending were to return to its record low as a share of GDP. While this is not necessarily an absolute floor, we believe it would be difficult for activity to breach these levels without a very large shock. Exhibit 5 summarizes the results.
We see limited further downside for investment in housing and nonresidential structures, even in a stressed scenario. Without a recovery, it is hard for these sectors to double-dip. These two components have contributed an average of -0.7pp to the 2.3% peak-to-trough decline in real GDP during post-war recessions (see again Exhibit 3). The fact that they have limited further downside is one major reason to think that a contraction today could be smaller.
Investment in equipment and software could contract more. Business equipment investment has recovered more and therefore has further to fall. According to our stress test, the potential drag from this component is about 1pp, or twice as much as in an average US recession. Consumer spending on durable goods also subtracts a small amount from GDP in our stress test, mostly reflecting declines in spending on goods other than motor vehicles (TVs, appliances, etc).
In total our stress scenario implies a drag of 1.6pp on growth if cyclical components of the economy fall back to record lows as a share of GDP. This estimate is close to the average drag from these components in earlier post-war recessions (1.7pp).
Adding it Up
To calculate the total impact on GDP we need estimates of the contribution from net exports and inventories (we take up government spending in the next section). Net exports depend on domestic demand, foreign growth, as well as relative import and export prices, but modeling these factors is beyond the scope of this article. Here we will assume that net exports simply respond to domestic final sales, and that the other factors affecting net exports behave similar to history. Historically, when the drag from domestic final sales has been around 1.6pp, the offset from net exports has been about 0.6-0.7pp. We will take the lower figure and assume that in a recession net exports would add 0.6pp to GDP growth.
For inventories we estimated a simple error-correction model that relates the level of private inventories to real final sales and a time trend (which captures the steady decline in inventory-to-sales ratios in recent decades). We then used this model to forecast the contribution to GDP growth from inventories. If final sales growth were -1%—a 1.6pp drag from consumption and investment in our “stress test” scenario less a 0.6pp boost from net exports—our model implies a GDP drag of 0.4pp from inventories. The drag from inventories is relatively low because 1) inventory levels are currently moderate compared to final sales, and so the correction would not need to be very large, and 2) the impulse from final sales is small due to low activity in cyclical sectors.
Putting it all together, we get a point estimate of -1.4% for the impact on GDP—smaller than the post-war average decline of -2.3%. This figure of course reflects our specific assumptions, and there is a wide range of possible outcomes (particularly for inventories). But it is encouraging that even if cyclical activity fell back to historic lows the result would likely be only an average recession. A very large recession therefore looks unlikely given the buffer of low cyclical activity.
Policy Constraints
Another important initial condition may cut the other way, however: if policymakers are constrained in some way and unable to respond to a recession, the contraction could be deeper or more prolonged than usual. To test this hypothesis we measured the stance of monetary and fiscal policy around previous business cycle turning points.
Exhibit 6 shows changes in the stance of fiscal policy around past recessions. The black bars represent “discretionary” policy, measured by changes in the cyclically-adjusted budget balance as a share of potential GDP. The gray bars represent the remaining cyclical portion of the federal budget, which we label “automatic stabilizers”.
The chart shows that in the year before GDP peaks fiscal policy has typically tightened modestly—through both discretionary policy changes and the impact of strong growth on automatic stabilizers. During contractions, fiscal policy eases sharply, but at first only because of the automatic stabilizers (unemployment insurance benefits, food stamps, the impact of progressive tax rates, etc). Discretionary fiscal policy usually does not kick in until later—in most cases until after the recovery has started.For the current outlook, this profile suggests that fiscal policy should still be expected to provide some support to GDP growth in a recession, because most of that support comes through automatic stabilizers.
The lack of room for additional discretionary fiscal policy does not necessarily mean a deeper recession, but it likely implies a tougher exit. For example, typical multipliers imply that the normal ease in discretionary fiscal policy two years after the peak adds around 1% to GDP in the first year of recovery. Without that support the post-recession rebound in GDP is likely to be more gradual.
Exhibit 7 shows the same analysis for changes in the federal funds rate. The pattern is broadly similar to that of fiscal policy changes, with two important differences. First, the Fed usually tightens more aggressively before recessions—consistent with the view that monetary policy tightening in response to inflation pressure is one common cause of recessions. Second, the Fed eases more quickly than discretionary fiscal policy.
Compared to fiscal policy, the implications for the present are more nuanced. On the one hand, the fact that monetary policy is not tightening means it is less likely to cause a recession. Shocks would need to come from other sources—possibly low confidence or fiscal consolidation. In this sense the stance of monetary policy could be a reason to expect a milder recession, because activity in earlier recessions was held back by the lagged effect of Fed tightening.
On the other hand, Fed policy is still “tight” compared to our Taylor Rule, so it is not obvious that monetary policy would be more supportive of growth today than in past cycles. Moreover, given that the Fed has already used its conventional policy tools and many of its unconventional tools as well, it may ease less aggressively in the event of a recession. Like fiscal policy, constrained monetary policy may be a reason to expect less rapid growth coming out of a downturn.
Standard multipliers suggest that the average 2.5pp cut in the funds rate during recessions would add 1.5% to GDP in the first year of recovery. Easing with unconventional tools could provide some boost to growth today, but is it likely that support from monetary policy would be smaller than after a typical recession.
Status of Financial System
A final factor affecting the severity of any future recession could be the state of the financial system. If financial distress returns it could lead to tighter financial conditions and a larger contraction in economic activity. We are cautiously optimistic that banking sector stress will not significantly amplify the recession as it did in 2008. As our bank analysts have emphasized, banks have better capital ratios, high loan loss reserves, more liquidity and less exposure to leveraged losses than in 2008. Better bank balance sheet health likely translates into more resilience during a downturn, and a smaller impact of weak activity on lending conditions. In addition, borrowing has fallen significantly and credit demand remains low. If the nonfinancial sector is already deleveraging, it may be less sensitive to tougher loan terms.
Spillovers from the European financial crisis and questions about how much policy support would be forthcoming in the event of future stress are sources of concern. But absent severe stress for global banks the healthier US financial system may also limit the potential downside for the economy.
Shallow Hole, Tougher Climb
Initial conditions will partly determine the shape of any future recession. Perversely, low activity in cyclical sectors of the economy is likely a major positive: residential investment, vehicle sales and spending on business structures are extremely low, and therefore less vulnerable to steep declines. All else equal, this suggests that an average size shock would lead to a relatively shallow recession today.
In contrast, constraints on policy—a focus on deficit reduction in Congress and maybe a perception that monetary policy has reached its limits—are a potential negative. While we do not think limited policy options would mean a deeper recession, they would likely imply a more tepid rebound.
Whatever the outcome, a US recession today would be painful. Given its high level, even a “small” increase could take the unemployment rate to 11-12%. If policy constraints mean that it will be difficult to bring unemployment back down, it is especially important to avoid a contraction in the first place.
India Equity Research Reports, IPO and Stock News
Visit http://indiaer.blogspot.com/ for complete details �� ��
Goldman Sachs Research
How Much Downside?
- Our base-case forecast continues to be that the US economy will avoid a recession, though the risks are high and rising.
- How bad could a recession be? Besides the size of shocks, we think three main factors would determine the depth of any downturn: 1) the level of activity in cyclical parts of the economy, 2) the stance of monetary and fiscal policy, and 3) the state of the financial system.
- Low cyclical activity and moderate inventory stocks are compelling reasons to expect a milder downturn. For example, current housing investment is mostly covering wear and tear. We estimate that in a “stress test” scenario, GDP would likely contract by about 1.5%—or less than in an average recession. A deep recession looks unlikely given already-low cyclical activity.
- Constraints on monetary and fiscal policy probably matter less for the depth of a recession than the strength of the subsequent recovery. Most fiscal easing during recessions comes through automatic stabilizers, and these would still provide support. However, discretionary fiscal stimulus can help quicken recoveries. Monetary policy eases faster than discretionary fiscal policy but is usually also tighter prior to recessions.
- The state of the financial system may also matter. Because of better bank balance sheet health, we are cautiously optimistic that financial stress would not significantly amplify a recession as it did in 2008.
- Given initial conditions, we see the main downside scenario as a shallow recession followed by a slow recovery. Low activity in cyclical sectors probably prevents a deep downturn, but a lack of policy room means it may be difficult to recover lost ground.
Exhibit 1 summarizes average outcomes during the eleven post-war US recessions. In a typical recession, real GDP contracts by 2.3% peak-to-trough. In the mildest US recession (2001), GDP fell by just 0.3% and actually rose over the year as a whole. In the most severe recession (2007-09) real GDP contracted by 5.1%. On average consumption has fallen by 0.6%, and nominal corporate profits by about 20%.
Recessions are also hard on the labor market. On average nonfarm payroll employment has declined by 2.8% peak-to-trough in the past. At the current level of employment, this would imply a loss of about 3.7 million jobs. In the mildest past recession for the labor market (1980), nonfarm payrolls fell by 0.9%—the equivalent of 1.2m jobs today. The unemployment rate typically rises by about 3 percentage points (pp) and has always risen by at least 1.9pp.
Any recession would clearly be harmful, but there have been a wide range of outcomes in the past. The severity of a recession would of course depend on the size of the shock hitting the economy—a large fiscal contraction would be worse than a small one, for example. But initial conditions probably matter as well. We think three main factors are important today: 1) the level of activity in cyclical parts of the economy, 2) the stance of monetary and fiscal policy, and 3) the state of the financial system.
Low Activity in Cyclical Sectors
The pullback in spending during recessions is typically concentrated in just a few segments of the economy: consumer spending on motor vehicles and other durable goods, residential investment, business fixed investment, and inventories. These highly cyclical sectors provide virtually all of the amplitude in GDP growth (Exhibits 2 & 3).
Because of the deep decline in output during the last recession and moderate recovery to date, activity in cyclical sectors remains relatively low. We argued last year that low levels of activity in the most cyclical parts of the economy reduce the odds of recession. In our view, recessions are partly corrections of imbalances that build up in the economy over time. If activity is still recovering from an earlier recession and those imbalances have not yet built up, growth may be more resilient to any given shock. Low activity in cyclical sectors is also likely to mitigate the severity of a potential recession: with spending still weak, it may be difficult to cut further in some areas.
The clearest example is housing. After declining by roughly 60% since its peak at the end of 2005, residential investment is only barely above the rate of depreciation in the housing capital stock (Exhibit 4). In other words, although housing investment has not fallen to zero, it is probably near its practical minimum as current investment is mostly just covering wear and tear. Despite ups and downs in the broader recovery, the main housing indicators—housing starts, new home sales, builder sentiment, etc.—have mostly remained flat at low levels, which also suggests the housing market has probably reached a floor. If the economy enters recession, we suspect that residential investment will not fall much further.
A similar argument likely holds for spending on motor vehicles. Vehicle sales are currently running at an annualized rate of just over 12 million units. This is close to the annual “scrappage” rate for vehicles in recent years, implying that recent sales merely keep the number of cars on the road constant. Lower sales are certainly possible, but from the current weak level this would require a smaller auto stock or a gradual deterioration in quality through depreciation.
Stress Test
To quantify the potential downside in cyclical components of final expenditure we estimated the impact on GDP if cyclical spending were to return to its record low as a share of GDP. While this is not necessarily an absolute floor, we believe it would be difficult for activity to breach these levels without a very large shock. Exhibit 5 summarizes the results.
We see limited further downside for investment in housing and nonresidential structures, even in a stressed scenario. Without a recovery, it is hard for these sectors to double-dip. These two components have contributed an average of -0.7pp to the 2.3% peak-to-trough decline in real GDP during post-war recessions (see again Exhibit 3). The fact that they have limited further downside is one major reason to think that a contraction today could be smaller.
Investment in equipment and software could contract more. Business equipment investment has recovered more and therefore has further to fall. According to our stress test, the potential drag from this component is about 1pp, or twice as much as in an average US recession. Consumer spending on durable goods also subtracts a small amount from GDP in our stress test, mostly reflecting declines in spending on goods other than motor vehicles (TVs, appliances, etc).
In total our stress scenario implies a drag of 1.6pp on growth if cyclical components of the economy fall back to record lows as a share of GDP. This estimate is close to the average drag from these components in earlier post-war recessions (1.7pp).
Adding it Up
To calculate the total impact on GDP we need estimates of the contribution from net exports and inventories (we take up government spending in the next section). Net exports depend on domestic demand, foreign growth, as well as relative import and export prices, but modeling these factors is beyond the scope of this article. Here we will assume that net exports simply respond to domestic final sales, and that the other factors affecting net exports behave similar to history. Historically, when the drag from domestic final sales has been around 1.6pp, the offset from net exports has been about 0.6-0.7pp. We will take the lower figure and assume that in a recession net exports would add 0.6pp to GDP growth.
For inventories we estimated a simple error-correction model that relates the level of private inventories to real final sales and a time trend (which captures the steady decline in inventory-to-sales ratios in recent decades). We then used this model to forecast the contribution to GDP growth from inventories. If final sales growth were -1%—a 1.6pp drag from consumption and investment in our “stress test” scenario less a 0.6pp boost from net exports—our model implies a GDP drag of 0.4pp from inventories. The drag from inventories is relatively low because 1) inventory levels are currently moderate compared to final sales, and so the correction would not need to be very large, and 2) the impulse from final sales is small due to low activity in cyclical sectors.
Putting it all together, we get a point estimate of -1.4% for the impact on GDP—smaller than the post-war average decline of -2.3%. This figure of course reflects our specific assumptions, and there is a wide range of possible outcomes (particularly for inventories). But it is encouraging that even if cyclical activity fell back to historic lows the result would likely be only an average recession. A very large recession therefore looks unlikely given the buffer of low cyclical activity.
Policy Constraints
Another important initial condition may cut the other way, however: if policymakers are constrained in some way and unable to respond to a recession, the contraction could be deeper or more prolonged than usual. To test this hypothesis we measured the stance of monetary and fiscal policy around previous business cycle turning points.
Exhibit 6 shows changes in the stance of fiscal policy around past recessions. The black bars represent “discretionary” policy, measured by changes in the cyclically-adjusted budget balance as a share of potential GDP. The gray bars represent the remaining cyclical portion of the federal budget, which we label “automatic stabilizers”.
The chart shows that in the year before GDP peaks fiscal policy has typically tightened modestly—through both discretionary policy changes and the impact of strong growth on automatic stabilizers. During contractions, fiscal policy eases sharply, but at first only because of the automatic stabilizers (unemployment insurance benefits, food stamps, the impact of progressive tax rates, etc). Discretionary fiscal policy usually does not kick in until later—in most cases until after the recovery has started.For the current outlook, this profile suggests that fiscal policy should still be expected to provide some support to GDP growth in a recession, because most of that support comes through automatic stabilizers.
The lack of room for additional discretionary fiscal policy does not necessarily mean a deeper recession, but it likely implies a tougher exit. For example, typical multipliers imply that the normal ease in discretionary fiscal policy two years after the peak adds around 1% to GDP in the first year of recovery. Without that support the post-recession rebound in GDP is likely to be more gradual.
Exhibit 7 shows the same analysis for changes in the federal funds rate. The pattern is broadly similar to that of fiscal policy changes, with two important differences. First, the Fed usually tightens more aggressively before recessions—consistent with the view that monetary policy tightening in response to inflation pressure is one common cause of recessions. Second, the Fed eases more quickly than discretionary fiscal policy.
Compared to fiscal policy, the implications for the present are more nuanced. On the one hand, the fact that monetary policy is not tightening means it is less likely to cause a recession. Shocks would need to come from other sources—possibly low confidence or fiscal consolidation. In this sense the stance of monetary policy could be a reason to expect a milder recession, because activity in earlier recessions was held back by the lagged effect of Fed tightening.
On the other hand, Fed policy is still “tight” compared to our Taylor Rule, so it is not obvious that monetary policy would be more supportive of growth today than in past cycles. Moreover, given that the Fed has already used its conventional policy tools and many of its unconventional tools as well, it may ease less aggressively in the event of a recession. Like fiscal policy, constrained monetary policy may be a reason to expect less rapid growth coming out of a downturn.
Standard multipliers suggest that the average 2.5pp cut in the funds rate during recessions would add 1.5% to GDP in the first year of recovery. Easing with unconventional tools could provide some boost to growth today, but is it likely that support from monetary policy would be smaller than after a typical recession.
Status of Financial System
A final factor affecting the severity of any future recession could be the state of the financial system. If financial distress returns it could lead to tighter financial conditions and a larger contraction in economic activity. We are cautiously optimistic that banking sector stress will not significantly amplify the recession as it did in 2008. As our bank analysts have emphasized, banks have better capital ratios, high loan loss reserves, more liquidity and less exposure to leveraged losses than in 2008. Better bank balance sheet health likely translates into more resilience during a downturn, and a smaller impact of weak activity on lending conditions. In addition, borrowing has fallen significantly and credit demand remains low. If the nonfinancial sector is already deleveraging, it may be less sensitive to tougher loan terms.
Spillovers from the European financial crisis and questions about how much policy support would be forthcoming in the event of future stress are sources of concern. But absent severe stress for global banks the healthier US financial system may also limit the potential downside for the economy.
Shallow Hole, Tougher Climb
Initial conditions will partly determine the shape of any future recession. Perversely, low activity in cyclical sectors of the economy is likely a major positive: residential investment, vehicle sales and spending on business structures are extremely low, and therefore less vulnerable to steep declines. All else equal, this suggests that an average size shock would lead to a relatively shallow recession today.
In contrast, constraints on policy—a focus on deficit reduction in Congress and maybe a perception that monetary policy has reached its limits—are a potential negative. While we do not think limited policy options would mean a deeper recession, they would likely imply a more tepid rebound.
Whatever the outcome, a US recession today would be painful. Given its high level, even a “small” increase could take the unemployment rate to 11-12%. If policy constraints mean that it will be difficult to bring unemployment back down, it is especially important to avoid a contraction in the first place.
No comments:
Post a Comment