Part 1: Recent Economic and Financial Developments
Monetary Policy Report submitted to the Congress on July 5, 2019, pursuant to section 2B of the Federal Reserve Act
The labor market strengthened further during the first half of 2019 but at a slower pace than last year...
Labor market conditions have continued to strengthen so far this year but at a pace slower than last year. Total nonfarm payroll employment has averaged gains of about 165,000 per month over the first five months of 2019, according to the Bureau of Labor Statistics. This pace is slower than the average monthly gains in 2018, but it is faster than what is needed to provide jobs for net new entrants into the labor force as the working-age population grows (figure 1).1
In April and May of this year, the unemployment rate stood at 3.6 percent, 1/4 percentage point lower than its level in December 2018 and its lowest level since 1969 (figure 2). In addition, the unemployment rate is 1/2 percentage point below the median of Federal Open Market Committee (FOMC) participants' estimates of its longer-run normal level.2
In May, the labor force participation rate (LFPR) for individuals 16 and over—that is, the share of the population either working or actively seeking work—was 62.8 percent, and it has changed little, on net, since late 2013. The aging of the population is an important contributor to an underlying downward trend in the overall participation rate. In particular, members of the baby-boom cohort are increasingly moving into their retirement years, ages when labor force participation typically falls. The flat trajectory in the overall LFPR is therefore consistent with strengthening labor market conditions; indeed, the LFPR for prime-age individuals (between 25 and 54 years old), which is much less sensitive to the effects of population aging, has been rising over the past few years (figure 3). Combining both the unemployment rate and the LFPR, the employment-to-population ratio (EPOP) for individuals 16 and over—the share of that segment of the population who are working—was 60.6 percent in May and has been gradually increasing throughout the expansion. The increase has been considerably larger for those with at least some college education than for those with no more than a high school diploma. (The box "How Have Lower-Educated Workers Fared since the Great Recession?" discusses movements in the EPOP by educational level over the current expansion.)
Other indicators are also consistent with strong labor market conditions. As reported in the Job Openings and Labor Turnover Survey (JOLTS), the average of the private-sector job openings rate over the first four months of the year was near its historical high, consistent with surveys indicating that businesses see vacancies as hard to fill. Similarly, the quits rate in the JOLTS is also near the top of its historical range, an indication that workers are being bid away from their current jobs or have become more confident that they can successfully switch jobs if they wish to. This interpretation accords well with surveys of consumers that indicate households perceive jobs as plentiful. The JOLTS layoff rate and the number of people filing initial claims for unemployment insurance benefits have both remained quite low.
How Have Lower-Educated Workers Fared since the Great Recession?
The U.S. labor market has been strengthening since the end of the Great Recession. Over this period, the unemployment rate has fallen roughly 6 percentage points, and the employment-to-population ratio (EPOP) for individuals between 25 and 54 years old (prime age) has increased about 4-1/4 percentage points. However, labor market outcomes during the expansion have been quite different for lower- and higher-educated individuals. The EPOP for prime-age college graduates declined about 2.5 percentage points during the recession, but it began a steady and sustained recovery in 2010 and was nearly at its pre-recession level by 2018 (left panel of figure A). In contrast, the EPOP for prime-age individuals with a high school degree or less fell much more sharply during the recession and lingered near its trough for several years before it began to recover in 2014.1 As of 2018, the EPOP for lower-educated workers remained well below its pre-recession level. In addition, real (or inflation-adjusted) hourly wages for lower-educated workers fell more over the 2007–13 period than real wages for college graduates (right panel of figure A). Real wages subsequently picked up for both groups, but cumulative real wage gains for lower-educated workers have only recently caught up, in percentage terms, to those for workers with college degrees.2
The relative underperformance of employment and wage growth for lower-educated workers has been a characteristic of all business cycles since at least 1980. This pattern is likely due, at least in part, to a long-term downward trend in the demand for lower-educated workers that is unrelated to the business cycle and caused, perhaps, by changes in technology and globalization.3 To focus on the effects of business cycles distinct from these longer-term trends, we examine business cycles at the state level to estimate the "typical" cyclical decline and recovery of employment for both education groups.
The typical state-level business cycle shows a starkly different evolution of employment for lower-educated workers compared with that for workers with college degrees. Typically in a recession, the EPOP declines immediately for both groups, but the decline is deeper and longer lasting for those with a high school degree or less (figure B).4 Once that group's EPOP begins a sustained recovery, though, it increases at a more rapid pace than the EPOP for those with a college degree. These results indicate that the EPOP for lower-educated workers may not fully recover for at least eight years, on average, following the end of a recession.
The differences in labor market outcomes over the business cycle for different education groups may in part be due to employers changing their hiring standards. Some research shows that employers raise skill requirements for new hires in a recession and then gradually lower skill requirements as the labor market recovers.5 Other research suggests that when high-skilled workers lose their jobs during recessions, they take jobs that require fewer skills, making these jobs less likely to be filled by low-skilled individuals.6 This pattern could at least in part explain the differences in labor market outcomes for lower- and higher-educated workers since the most recent recession.
As the unemployment rate falls and employers relax their hiring standards, more opportunities are likely to open for lower-educated workers. Aaronson and others (2019) present some evidence that disadvantaged groups, such as nonwhite individuals and those with less education, benefit more from further improvement in the labor market relative to more advantaged groups when the unemployment rate is below its natural rate.7 Indeed, real wages for lower-educated workers rose faster over the past few years as the labor market tightened, and total wage growth for those workers since 2007 is now close to wage growth for more-educated workers (as shown in the right panel of figure A). Hotchkiss and Moore (2018) find that exposure to a low-unemployment economy is particularly beneficial for individuals who entered the labor market during periods of high unemployment and would otherwise face persistently worse labor market outcomes.8 Thus, periods of low unemployment may particularly improve labor market outcomes of lower-educated workers.
1. The analysis excludes those with some college education but not a four-year degree. The labor market experience of such individuals, though, is similar to that of individuals with a high school degree or less. Return to text
2. Another measure of wage growth using the same Current Population Survey data source is the Federal Reserve Bank of Atlanta's Wage Growth Tracker (WGT), which calculates the median, year-over-year percent change in nominal wages of individuals employed 12 months apart. The WGT measure shows that median wage growth for workers with a high school degree was lower than median wage growth for workers with a college degree through 2015. Since then, median wage growth for both groups has been similar. Return to text
3. The EPOP for lower-educated, prime-age individuals has been trending lower for men since 1950 and for women since 2000, largely reflecting the trends in those groups' labor force participation rates (LFPRs). For an overview of factors affecting the LFPRs of prime-age individuals, see the box "The Labor Force Participation Rate for Prime-Age Individuals" in Board of Governors of the Federal Reserve System (2018), Monetary Policy Report (Washington: Board of Governors, July), pp. 8–10, https://www.federalreserve.gov/monetarypolicy/files/20180713_mprfullreport.pdf; and Congressional Budget Office (2018), Factors Affecting the Labor Force Participation of People Ages 25 to 54 (Washington: CBO, February, https://www.cbo.gov/system/files/115th-congress-2017-2018/reports/53452-lfpr.pdf). Return to text
4. For ease of interpretation, we define a typical recession as a state experiencing a temporary 1 percent decline in state output growth in a given year, returning to normal growth in the following year. To get the estimated effect of a larger or smaller recession, simply multiply the estimates by the specified decline in output growth. Return to text
5. See Brad Hershbein and Lisa B. Kahn (2018), "Do Recessions Accelerate Routine-Biased Technological Change? Evidence from Vacancy Postings," American Economic Review, vol. 108 (July), pp. 1737–72; and Alicia Sasser Modestino, Daniel Shoag, and Joshua Ballance (2016), "Downskilling: Changes in Employer Skill Requirements over the Business Cycle," Labour Economics, vol. 41 (August), pp. 333–47. Return to text
6. See Regis Barnichon and Yanos Zylberberg (2019), "Underemployment and the Trickle-Down of Unemployment," American Economic Journal: Macroeconomics, vol. 11 (April), pp. 40–78. Return to text
7. See Stephanie R. Aaronson, Mary C. Daly, William Wascher, and David W. Wilcox (2019), "Okun Revisited: Who Benefits Most from a Strong Economy?" paper presented at the Brookings Papers on Economic Activity Conference, held at the Brookings Institution, Washington, March 7–8, https://www.brookings.edu/wp-content/uploads/2019/03/Okun-Revisited-Who-Benefits-Most-From-a-Strong-Economy.pdf. Return to text
8. See Julie L. Hotchkiss and Robert E. Moore (2018), "Some Like It Hot: Assessing Longer-Term Labor Market Benefits from a High-Pressure Economy," Andrew Young School of Policy Studies Research Paper Series 18-01 (Atlanta: Georgia State University, February), https://aysps.gsu.edu/files/2018/03/18-01-HotchkissMoore-SomelikeitHot.pdf. Return to textReturn to text
. . . and unemployment rates have fallen for all major demographic groups over the past several years
Differences in unemployment rates across ethnic and racial groups have narrowed in recent years, as they typically do during economic expansions, after having widened during the recession (figure 4). However, unemployment rates for African Americans and Hispanics remain substantially above those for whites and Asians. The rise in LFPRs for prime-age individuals over the past few years has also been apparent in each of these racial and ethnic groups (figure 5).
Increases in labor compensation have picked up but remain moderate by historical standards...
Despite strong labor market conditions, the available indicators generally suggest that increases in hourly labor compensation have remained moderate. The employment cost index—a measure of both wages and the cost to employers of providing benefits—was 2-3/4 percent higher in March of 2019 relative to its year-earlier level (figure 6). Compensation per hour in the business sector—a broad-based but volatile measure of wages, salaries, and benefits—rose 1-1/2 percent over the four quarters ending in 2019:Q1, less than the annual increases over the preceding couple of years. Among measures that do not account for benefits, average hourly earnings rose 3.1 percent in May relative to 12 months earlier, a slightly faster rate of increase than during the same period of a year ago. According to the Federal Reserve Bank of Atlanta, the median 12-month wage growth of individuals reporting to the Current Population Survey increased about 3-3/4 percent in May, near the upper portion of its range over the past couple of years.3
. . . and likely have been restrained by slow growth in labor productivity over much of the expansion
These moderate rates of hourly compensation gains likely reflect the offsetting influences of a strengthening labor market and productivity growth that has been weak through much of the expansion. From 2008 to 2017, labor productivity increased a little more than 1 percent per year, on average, well below the average pace from 1996 to 2007 of nearly 3 percent and also below the average gain in the 1974–95 period (figure 7). Although considerable debate remains about the reasons for the slowdown in productivity growth over this period, the weakness may be partly attributable to the sharp pullback in capital investment during the most recent recession and the relatively slow recovery that followed. More recently, however, labor productivity rose 1-3/4 percent in 2018 and picked up further in the first quarter of 2019.4 While it is uncertain whether this faster rate of growth will persist, a sustained pickup in productivity growth, as well as additional labor market strengthening, would support stronger gains in labor compensation.
Price inflation has dipped below 2 percent this year
Consumer price inflation has moved down below the FOMC's objective of 2 percent this year.5 As measured by the 12-month change in the price index for personal consumption expenditures (PCE), inflation is estimated to have been 1.5 percent in May after being at or above 2 percent for much of 2018 (figure 8). Core PCE inflation—which excludes consumer food and energy prices that are often quite volatile, and which therefore typically provides a better indication than the total measure of where overall inflation will be in the future—also moved lower in recent months and is estimated to have been 1.6 percent over the 12 months ending in May. The slowing in core inflation to date reflects particularly low readings in the first three months of the year that appear due to idiosyncratic price declines in a number of specific categories such as apparel, used cars, and banking services and portfolio management services. Indeed, in April and May, core inflation accelerated, posting larger average monthly gains than in the first quarter.
The trimmed mean PCE price index, produced by the Federal Reserve Bank of Dallas, provides an alternative way to purge inflation of transitory influences, and it is less sensitive than the core index to idiosyncratic price movements such as those noted earlier.6 The 12-month change in this measure was 2 percent in May.
Oil prices rebounded through the spring but have moved down recently...
After dropping sharply late last year, the Brent price of crude oil moved up to almost $75 per barrel in mid-April, partly reflecting declines in production in Iran and Venezuela and voluntary supply cuts by other OPEC members and partner countries (figure 9). More recently, however, prices have fallen back to around $65 per barrel because of concerns about global growth. The changes in oil prices have contributed to similar movements in retail gasoline prices, which rose through early spring but have also fallen back recently.
. . . and prices of imports other than energy fell
Nonfuel import prices, before accounting for the effects of tariffs on the price of imported goods, have continued to decline from their mid-2018 peak, responding to dollar appreciation, lower foreign inflation, and declines in non-oil commodity prices (figure 10).7 In particular, prices of industrial metals have fallen in recent months, partly on concerns about weak global demand.
Survey-based measures of inflation expectations have been stable...
Expectations of inflation likely influence actual inflation by affecting wage- and price-setting decisions. Survey-based measures of inflation expectations at medium- and longer-term horizons have remained generally stable over the past year. In the Survey of Professional Forecasters, conducted by the Federal Reserve Bank of Philadelphia, the median expectation for the annual rate of increase in the PCE price index over the next 10 years has been very close to 2 percent for the past several years (figure 11). In the University of Michigan Surveys of Consumers, the median value for inflation expectations over the next 5 to 10 years has fluctuated around 2-1/2 percent since the end of 2016, though this level is about 1/4 percentage point lower than had prevailed through 2014. In the Survey of Consumer Expectations, conducted by the Federal Reserve Bank of New York, the median of respondents' expected inflation rate three years hence has fluctuated between 2-1/2 percent and 3 percent over the past five years.
. . . while market-based measures of inflation compensation have come down since the first half of 2018
Inflation expectations can also be gauged by market-based measures of inflation compensation. However, the inference is not straightforward, because market-based measures can be importantly affected by changes in premiums that provide compensation for bearing inflation and liquidity risks. Measures of longer-term inflation compensation—derived either from differences between yields on nominal Treasury securities and those on comparable-maturity Treasury Inflation-Protected Securities (TIPS) or from inflation swaps—tend to fall when markets are volatile because of the incorporation of liquidity risks. Such declines occurred around the turn of the year and again in May and June, when market volatility picked up again. Despite the fluctuations this year, these measures of inflation compensation remain notably below levels that prevailed in the summer of 2018 (figure 12).8 The TIPS-based measure of 5-to-10-year-forward inflation compensation and the analogous measure from inflation swaps are now about 1-3/4 percent and 2 percent, respectively, with both measures below their respective ranges that prevailed for most of the 10 years before the start of the notable declines in mid-2014.9
Real gross domestic product growth was strong in the first quarter, but there are recent signs of slowing
Real gross domestic product (GDP) rose at an annual rate of 3 percent in 2018 (figure 13). In the first quarter, real GDP again moved up at an annual rate of around 3 percent. However, there are indications that growth will moderate in the second quarter.10 Net exports and business inventories provided a sizable boost to first-quarter GDP growth, but their contributions appear to have reversed in the months following. Notably, private domestic final purchases—that is, final purchases by households and businesses, which tend to provide a better indication of future GDP growth than most other components of overall spending—posted only a modest increase in the first quarter. The slowing that occurred in consumer spending appears to have been temporary, but the slowing in business fixed investment appears to be more persistent. Manufacturing output fell in the first quarter, and it moved down further in April before posting a small gain in May. Although lower production levels of motor vehicles and aircraft were important contributors to the weakness, the recent declines in manufacturing were broad based.11 Nevertheless, the economic expansion continues to be abetted by steady job gains, increases in household wealth, expansionary fiscal policy, and still-supportive domestic financial conditions, including moderate borrowing costs and easy access to credit for many households and businesses.
Growth in business fixed investment has softened after strong gains in 2018
Investment spending by businesses rose rapidly in 2018 but appears to have decelerated sharply this year. In the first quarter, growth slowed to an annual rate of 4-1/2 percent, while new orders for nondefense capital goods, excluding the volatile aircraft category, have declined modestly, on balance, in recent months (figure 14). In addition, forward-looking indicators of business spending such as capital spending plans have deteriorated amid downbeat business sentiment and profit expectations from industry analysts, reportedly reflecting trade tensions and concerns about global growth.
By contrast, activity in the housing sector had been declining but recently shows signs of stabilizing
Residential investment fell in 2018 and declined further in the first quarter. More recently, the pace of construction activity appears to have stabilized as housing starts for single-family and multifamily housing units rose, on average, in April and May (figure 15). Existing home sales moved higher as well over the same period, while new home sales moved down a bit following a sizable increase in the first quarter (figure 16). Consumers' perceptions of homebuying conditions and housing affordability have improved, which is consistent with the declines in mortgage rates this year and the slowing in growth of home prices (figure 17).
Ongoing improvements in the labor market and gains in wealth continue to support household income and consumer spending...
After increasing at a moderate pace of 2-1/2 percent in 2018 as a whole, real consumer spending slowed considerably in the first quarter (figure 18). However, incoming data suggest that consumer spending picked up in recent months, with PCE in May up at an annual rate of 2-1/2 percent relative to the average level in the fourth quarter.
Real disposable personal income (DPI), a measure of households' after-tax purchasing power, increased at a solid annual rate of 3 percent in 2018; however, so far this year, growth in real DPI has been more moderate despite strong gains in wage and salary income. With consumer spending rising more than disposable income so far this year, the personal saving rate moved down from an average of 6-1/2 percent in the fourth quarter to around6 percent in May (figure 19).
Ongoing gains in household wealth have likely continued to support consumer spending. House prices, which are of particular importance for the balance sheet positions of a large portion of households, continued to increase through May, although at a more moderate pace than in recent years (figure 20). In addition, U.S. equity prices, which fell sharply at the end of 2018, have rebounded this year. Buoyed by increases in home and equity prices, aggregate household net worth moved up to 6.8 times household income in the first quarter (figure 21).
. . . and consumer sentiment remains strong
Consumers have remained upbeat. Although the Michigan index of consumer sentiment dipped at the turn of the year, it has since rallied, and the sentiment measure from the Conference Board survey also climbed in the second quarter from its first-quarter level (figure 22). In June, both the Michigan and the Conference Board indexes of consumer sentiment were about in the middle of their ranges over the past few years.
Borrowing conditions for households remain generally favorable...
Despite increases in interest rates for consumer loans and some reported further tightening in credit card lending standards, financing conditions for consumers largely remain supportive of growth in household spending. Consumer credit expanded at a moderate pace in the first quarter, rising faster than disposable income (figure 23). Mortgage credit has continued to be readily available for households with solid credit profiles but remains noticeably tighter than before the most recent recession for borrowers with low credit scores. Standards for automotive loans have been generally stable, and overall delinquency rates for these loans were little changed in the first quarter at a moderate level. Financing conditions in the student loan market remain firm, with over 90 percent of such credit being extended by the federal government. After peaking in 2013, delinquencies on such loans have been gradually declining, reflecting in part the continued improvements in the labor market.
. . . while corporate financing conditions tightened somewhat relative to last year but remained accommodative overall
Aggregate flows of credit to large nonfinancial firms remained strong in the first quarter, supported in part by relatively low interest rates and accommodative financing conditions (figure 24). The gross issuance of corporate bonds, which had fallen substantially in December, rebounded in the first quarter as market volatility receded. After increasing notably in late 2018, spreads on both investment- and speculative-grade corporate bonds over comparable-maturity Treasury securities have both declined, on net, this year as investors' risk appetite seems to have recovered. In April, respondents to the Senior Loan Officer Opinion Survey on Bank Lending Practices, or SLOOS, reported that demand for commercial and industrial loans weakened in the first quarter even as lending standards remained unchanged and terms for such loans eased.12 However, banks reported tightening lending standards on all categories of commercial real estate loans. Meanwhile, financing conditions for small businesses have remained generally accommodative, but credit growth has been subdued.
Net exports supported GDP growth in the first quarter
After being a small drag on U.S. real GDP growth last year, net exports, which can have sizable swings from quarter to quarter, added about 1 percentage point to the rate of growth in the first quarter. Real U.S. exports increased at an annual rate of about 5-1/2 percent, as exports of agricultural products and automobiles expanded robustly. Real imports fell 2 percent following solid increases in 2018 (figure 25). Nominal goods trade data through May suggest that exports edged down in the second quarter, while imports were about flat. The available data suggest that the trade deficit and the current account in the first half of the year were little changed as a percent of GDP from 2018 (figure 26).
Federal fiscal policy actions boosted economic growth in 2018 but had a smaller effect on first-quarter real GDP because of the partial government shutdown...
Fiscal policy at the federal level boosted GDP growth in 2018 because of lower personal and business income taxes from the Tax Cuts and Jobs Act of 2017 and because of an increase in federal purchases due to the Bipartisan Budget Act of 2018.13 After increasing 2-3/4 percent in 2018, federal government purchases were flat in the first quarter of 2019, reflecting the effects of the partial federal government shutdown (figure 27). The government shutdown, which was in effect from December 22 through January 25, held down GDP growth in the first quarter, largely because of the lost work of furloughed federal government workers and affected federal contractors. That said, federal purchases are expected to rebound in the second quarter.
The federal unified budget deficit widened in fiscal year 2018 to around 4 percent of nominal GDP from 3-1/2 percent of GDP in 2017 because receipts moved lower, to 16 percent of GDP (figure 28). Expenditures are currently around 21 percent of GDP, slightly above the level that prevailed in the decade before the start of the 2007–09 recession. The ratio of federal debt held by the public to nominal GDP rose to around 77 percent in fiscal 2018 and was quite elevated relative to historical norms (figure 29). The Congressional Budget Office projects that this ratio will rise further over the next several years.
. . . and the fiscal position of most state and local governments is stable
The fiscal position of most state and local governments is stable, although there is a range of experiences across these governments. The revenue of state governments has grown moderately in recent quarters, as the economic expansion continues to push up income and sales tax collections. At the local level, property tax collections continue to rise, pushed higher by past house price gains. Real state and local government purchases grew modestly last year; however, outlays have surged so far this year, driven largely by a boost in construction spending. State and local infrastructure spending was weak for many years, and there appears to be demand for higher expenditures in this area. State and local government payrolls expanded slowly last year and over the first five months of 2019, and employment by these governments remains below its peak before the current expansion.
The expected path of the federal funds rate over the next several years has moved down
Market-based measures of the expected path for the federal funds rate over the next several years have declined substantially since the end of 2018 (figure 30). Various factors contributed to this shift, including increased investor concerns about downside risks to the global economic outlook and rising trade tensions. In addition, investors reportedly interpreted FOMC communications over the first half of 2019 as signaling the Federal Reserve is likely to lower the target range for the federal funds rate in light of muted inflation pressures and uncertainties about the global economic outlook.
Survey-based measures of the expected path of the policy rate also shifted down relative to the levels observed at the end of 2018. According to the results of the most recent Survey of Primary Dealers and Survey of Market Participants, both conducted by the Federal Reserve Bank of New York just before the June FOMC meeting, the median of respondents' modal projections implies a declining trajectory for the target range of the federal funds rate for 2019, which flattens out in 2020. Relative to the December survey, the median of these projections moved down 50 basis points for July 2019 and 100 basis points for December 2019.14 Additionally, market-based measures of uncertainty about the policy rate approximately one to two years ahead increased, on balance, from their levels at the end of last December.
The nominal Treasury yield curve has moved down and continued to flatten
Since the end of 2018, the nominal Treasury yield curve shifted down and flattened further, with the 2-, 5-, and 10-year nominal Treasury yields all declining about 70 basis points on net (figure 31). The decrease in Treasury yields, which is consistent with the revision in market participants' expectations for the path of policy rates, largely reflects FOMC communications as well as investors' concerns about the global economic outlook and the escalation of trade disputes. Option-implied volatility on swap rates—an indicator of uncertainty about Treasury yields—has increased notably, on net, since the beginning of the year. In particular, measures of near-term interest rate uncertainty have reached the levels seen at the end of 2018.
Yields on 30-year agency mortgage-backed securities (MBS)—an important factor influencing mortgage interest rates—decreased in line with the decline in the 10-year nominal Treasury yield and remained low by historical standards (figure 32). Likewise, yields on both investment-grade and high-yield corporate debt declined significantly from the levels in late 2018 and stayed very low (figure 33). Despite widening in May, the spreads on corporate bond yields over comparable-maturity Treasury yields have narrowed, on net, over the first half of 2019 and are close to their historical medians.
Broad equity price indexes increased on net
After declining sharply at the end of 2018, broad U.S. stock market indexes have recovered, on net, over the first half of 2019 (figure 34). The broad rebound in stock prices—which included all major economic sectors—was reportedly supported by Federal Reserve communications that were perceived as more accommodative than previously anticipated. Stocks fluctuated in May and June as downside risks and trade tensions were offset by further expectations of easier monetary policy.
Measures of implied and realized stock price volatility for the S&P 500 index declined notably on net. Following the highs seen at the end of 2018, these volatility measures declined until late April, with the VIX—a measure of implied volatility—returning to near the 10th percentile of its historical distribution and with realized volatility close to the 30th percentile of its historical range (figure 35). At the beginning of May, following the escalation of trade tensions, these volatility measures increased and have remained elevated since then, but they have stayed well below the high levels of December and now stand close to their historical medians. Several measures of financial conditions that aggregate large sets of financial data into summary indexes eased considerably since the end of 2018 but have tightened a bit since the beginning of May, in line with the decline in stock prices over that month, and have remained relatively elevated since then. (For a discussion of financial stability issues, see the box "Developments Related to Financial Stability.")
Developments Related to Financial Stability
The framework used by the Federal Reserve Board for assessing the resilience of the U.S. financial system focuses on financial vulnerabilities in four broad areas: asset valuations, household and business debt, leverage in the financial sector, and funding risks. The Financial Stability Report published on May 6, 2019, presents the most recent, detailed assessment of these vulnerabilities.1 This discussion summarizes its key findings, updated, where appropriate, to reflect developments since its publication.
Asset valuations remain somewhat elevated in a number of markets. Treasury term premiums are near record lows. Forward-looking measures of Treasury market volatility have recently increased, especially for shorter-dated Treasury securities. Equity prices appear to be somewhat elevated relative to earnings, with the forward equity price-to-earnings ratio for the S&P 500 remaining above the median value of its historical distribution since the mid-1980s (figure A). In commercial real estate markets, capitalization rates remain at historically low levels. Residential real estate prices are also somewhat high relative to rents (accounting for borrowing costs and long-run trends), although house price growth slowed materially in the past year. Valuation pressures in the leveraged loan market eased somewhat in recent months, and the spreads on lower-rated leveraged loans are now above the median value over the past 20 years. In corporate bond markets, spreads of 10-year corporate bonds over benchmark rates are close to the median of their historical distributions.
Vulnerabilities associated with total private-sector credit remain at a moderate level relative to the past several decades, and total debt has advanced roughly in line with economic activity over the past five years. Leverage in the business sector remains near its highest level in the past 20 years, and business debt has grown faster than gross domestic product (GDP) since 2012 (figure B). Rapid debt growth, while broad based across different parts of the business sector, is concentrated among the riskiest firms, and there are signs that credit standards for new leveraged loans are weak and have deteriorated further over the past six months. In the corporate bond market, the distribution of credit ratings among investment-grade bonds has worsened, with the share of bonds rated Baa (or triple-B) reaching near-record levels. While broader corporate credit performance remains solid amid a growing economy and debt-service costs are relatively low, a broader repricing of risk or a slowdown in economic activity could pose notable risks to borrowing firms and their creditors. Such developments could increase the downside risk to economic activity more generally. In contrast, in the household sector, debt growth is concentrated among borrowers with high credit scores, and the debt-to-GDP ratio continues to trend down (figure B).
Vulnerabilities stemming from leverage at financial institutions remain low. Capital relative to risk-weighted assets at the largest banks has remained largely stable over the past few years. Results of the annual Dodd-Frank Act Stress Tests, released on June 21, 2019, indicate that participating banks are sufficiently resilient to continue lending to creditworthy borrowers even in a severe macroeconomic scenario. The exposure of banks to nonbank financial institutions—such as finance companies, asset managers, securitization vehicles, and mortgage real estate investment trusts—continued to increase in the first quarter of 2019. Some of those firms are significant business lenders, adding to banks' exposure to elevated losses in the corporate sector. Leverage of broker-dealers increased slightly in 2018 but remains near historically low levels. Leverage has also stayed low at life insurance companies and at property and casualty insurance firms. At hedge funds, leverage increased in the first quarter of 2019 to levels slightly below its 2018 peak.
Vulnerabilities stemming from liquidity and maturity mismatches remain low. Banks hold large quantities of liquid assets, and their reliance on short-term wholesale funding is near its historical lows. Although assets under management at prime money market funds—which are more susceptible to runs than government funds—have increased since the U.S. Securities and Exchange Commission (SEC) reforms went into place in 2016, they remain well below their pre-reform levels. Holdings of U.S. corporate bonds by mutual funds increased substantially over the past decade, raising concerns about the mismatch between daily redemptions allowed by these funds and the time required to sell their assets. Rules adopted in 2016 by the SEC to strengthen mutual funds' and exchange-traded funds' liquidity risk management have started going into effect in the past year.2
Downside risks to U.S. financial stability from abroad remain moderate, but several near-term risk events could generate meaningful spillovers to the United States. Two prominent European risks are a "no deal" Brexit, which remains a possible outcome later in the year, and Italian fiscal challenges. Also, an escalation of the trade tensions between the United States and its major trading partners, along with financial market reactions, could exacerbate uncertainty and increase the downside risk to global economic activity. In China, high levels of nonfinancial-sector debt expose the financial sector to a slowdown in economic growth. The effects of any of these events on global financial markets could be amplified if they deepen the stresses in already vulnerable emerging market economies. These dynamics could tighten financial conditions in the United States and negatively affect the creditworthiness of U.S. firms.
The countercyclical capital buffer (CCyB) is a macroprudential tool that the Federal Reserve can use to increase the resilience of the financial system by raising capital requirements on internationally active banking organizations when financial vulnerabilities are meaningfully above normal. On March 6, 2019, the Board voted to maintain the CCyB at 0 percent.
Markets for Treasury securities, mortgage-backed securities, and municipal bonds have functioned well
Available indicators of Treasury market functioning have generally remained stable since the beginning of 2019, with a variety of measures—including bid-ask spreads, bid sizes, and estimates of transaction costs—displaying few signs of liquidity pressures. Liquidity conditions in the agency MBS market were also generally stable. Credit conditions in municipal bond markets remained stable as well, with yield spreads on 20‑year general obligation municipal bonds over comparable-maturity Treasury securities declining somewhat on net.
Money market rates were little changed
Rates across money markets were little changed, on balance, in the first half of 2019. Conditions in domestic short-term funding markets continued to be broadly stable since the end of 2018. Overnight secured and unsecured rates declined in line with the technical adjustment announced after the May FOMC meeting, which lowered the interests paid on required and excess reserve balances by 5 basis points. Other short-term interest rates, including those on commercial paper and negotiable certificates of deposit, were also little changed since the beginning of the year.
Bank credit continued to expand, and bank profitability remained robust
Credit provided by commercial banks to fund businesses as well as commercial and residential real estate continued to grow in 2019, albeit at a slower pace than in the second half of 2018. By contrast, consumer loan growth accelerated since the beginning of the year. In the first quarter of 2019, the pace of total bank credit expansion was about in line with that of nominal GDP, leaving the ratio of total commercial bank credit to current-dollar GDP little changed relative to last December (figure 36). Overall, measures of bank profitability remained solid in the first quarter of 2019, supported by wider net interest margins and steady loan growth (figure 37).
Advanced foreign economies have been slowly emerging from the recent soft patch
After a significant slowdown in the second half of last year, growth picked up in many advanced foreign economies (AFEs) at the start of 2019, but at a still restrained pace (figure 38). Notwithstanding continued weakness in the manufacturing sector and softening external demand, domestic demand in the AFEs generally improved amid rising employment and wages as well as easier financial conditions. The pickup in growth also reflected temporary factors. Economic activity in the euro area was boosted by the fading effects of car production disruptions in Germany and protests in France in 2018. Growth in the United Kingdom surged as expectations of trade disruptions surrounding the original date of the United Kingdom's exit from the European Union, or Brexit, led to stockpiling by households and firms. Economic activity in Canada, by contrast, remained depressed by oil production cuts, but recent indicators point to a rebound in growth in the second quarter.
Core inflation remained low in advanced foreign economies
The rebound in energy prices earlier in the year pushed up consumer price inflation in many AFEs (figure 39). However, despite further improvement in labor market conditions, inflationary pressures remained contained, with core inflation readings notably muted in the euro area and Japan. In Canada and the United Kingdom, by contrast, core inflation rates moved close to 2 percent.
AFE central banks took a more accommodative policy stance
With activity only slowly picking up and core inflation persistently low, European Central Bank (ECB) communications took a more accommodative tone. In March, the ECB indicated that it would keep its policy rate in negative territory through at least the middle of next year and rolled out a new round of loans for euro-area banks to reduce the risk of renewed funding pressures. In June, ECB President Mario Draghi added that the ECB would introduce new stimulus measures if the economic outlook did not improve. The Bank of Canada and Bank of England signaled more-gradual increases in interest rates, given a moderation in the pace of global economic activity. The Reserve Bank of Australia in June and July cut its policy rate in response to below-target inflation and weak economic growth.
Central banks' more accommodative policy stances supported AFE asset prices
The more accommodative policy stance in major AFEs contributed to an overall easing of financial conditions in the first half of the year. Market-implied paths of policy rates and long-term interest rates on sovereign bonds have generally fallen sharply, as in the United States (figure 40). Broad stock market indexes across AFEs are up, on net, since January (figure 41). However, concerns about global growth and rising trade tensions weighed on risky asset prices over the course of May and June. Sovereign bond spreads in Italy fluctuated amid uncertainty about the country's fiscal outlook.
Economic activity in emerging Asia struggled to gain a solid footing
In China, real GDP growth picked up in the first quarter, supported in part by fiscal and monetary policy measures that targeted smaller businesses and infrastructure spending, as well as by the more favorable financial conditions amid investor optimism on a U.S.–China trade deal (figure 42). Recent activity indicators, however, suggest that the underlying momentum in the economy remains relatively subdued against the backdrop of reemerging trade tensions, global weakness in trade and manufacturing, and the Chinese authorities' continued caution about providing substantial further credit stimulus. Amid moderating global trade and activity, real GDP growth in other Asian economies in the first quarter generally remained below their 2018 pace, with growth in Korea turning negative (see the box "The Persistent Slowdown in Global Trade and Manufacturing").
The Persistent Slowdown in Global Trade and Manufacturing
After expanding briskly in 2017, the growth of global goods trade and manufacturing, as indicated by industrial production, has slowed significantly (figure A). Even so, other aspects of economic activity, importantly services, have held up. A number of factors are likely contributing to the recent slowdown in trade and manufacturing growth, and disentangling them is difficult. First, new tariffs appear to have lowered imports and exports in the United States and elsewhere, while uncertainty surrounding trade policy could be leading firms to delay investment decisions and reduce capital expenditures. Second, a downturn in global sales for technology goods has restrained trade and manufacturing activity, especially in emerging Asia. Finally, a general slowdown in global demand, reflecting idiosyncratic factors specific to different economies, has likely weighed on demand for traded goods.
Regarding the first of these factors, global trade tensions have risen sharply since early 2018, fueled by both higher tariffs and uncertainty about the prospects for future trade policy. The United States has increased tariffs on over $250 billion in imports by an average of nearly 25 percentage points, and U.S. trading partners have retaliated. Several studies indicate that most of the cost of these higher tariffs has been passed through to U.S. importers.1 If we assume a commonly used elasticity of 1.5 for the response of imports to changes in prices, it implies that tariffs may have lowered U.S. imports by about $70 billion, or about 0.5 percent of world goods imports.2 Taking into account the effect of higher tariffs imposed by foreign countries as well, these estimates suggest that the overall direct effects of higher tariffs on global trade flows are, to date, likely to be material but modest relative to the observed step-down from 5.7 percent growth in 2017 to 1.5 percent growth in 2018.
In addition to the direct effect of the tariffs, however, rising uncertainty about the prospects for trade policy may also be weighing on trade and manufacturing. Measures of trade policy uncertainty spiked last year, largely reflecting concerns about current and prospective tariff hikes along with renegotiations of trade agreements (figure B). Higher uncertainty may lead businesses to delay investment purchases as they wait for the policy uncertainties to be resolved. Indeed, investment spending growth has slowed in many areas of the global economy since 2017. Although this slowdown may reflect a number of factors, concerns about trade policy have been flagged in many recent surveys of business attitudes and intentions, including the Beige Book.
The global tech cycle—a synchronized pattern of production and trade in electronics and software across economies—has also contributed to the decline in global trade and manufacturing growth. This cycle is particularly important for emerging Asia, where about one-third of exports are technology related. Global semiconductor sales surged in 2017 but fell sharply in the last months of 2018 (figure C). The fall in large part reflected a contraction in demand in China, particularly evident in mobile phone purchases. Recent data, however, suggest that this cycle may have bottomed out, as Chinese mobile phone production picked up in April along with exports of electronics in emerging Asia through May.
Finally, a regular feature of the data is that trade and manufacturing production move with overall gross domestic product (GDP) growth but with considerably more cyclical volatility (a pattern that can be seen in figure A). Trade and manufacturing production largely consist of durable goods, the purchase of which tends to be especially sensitive to economic conditions. Thus, although global trade and manufacturing slowed much more sharply than GDP last year, part of their sharp slowing likely just reflected a response to a more general slowing in global economic growth. A number of factors have contributed to the step-down in global activity. A deleveraging campaign by China's authorities was an important factor in the slowdown of the Chinese economy. Growth in Europe has been restrained by complications with meeting tighter emissions standards for new motor vehicles in Germany, protests in France, and the ongoing uncertainties associated with Brexit. And financial stresses have weighed on some emerging market economies, especially Argentina and Turkey.
1. For two recent working papers that analyze the effects of the tariff changes on trade volumes and import prices, see Mary Amiti, Stephen J. Redding, and David E. Weinstein (2019), "The Impact of the 2018 Trade War on U.S. Prices and Welfare," CEPR Discussion Paper DP13564 (London: Centre for Economic Policy Research, March), https://cepr.org/sites/default/files/news/FreeDP_Mar05.pdf; and Pablo D. Fajgelbaum, Pinelopi K. Goldberg, Patrick J. Kennedy, and Amit K. Khandelwal (2019), "The Return to Protectionism," NBER Working Paper Series 25638 (Cambridge, Mass.: National Bureau of Economic Research, March). Return to text
2. See David K. Backus, Patrick J. Kehoe, and Finn E. Kydland (1994), "Dynamics of the Trade Balance and the Terms of Trade: The J-Curve?" American Economic Review, vol. 84 (March), pp. 84–103. Return to textReturn to text
Latin American economies continued to underperform
In Mexico, real GDP contracted in the first quarter following generally weak performance in the past two years. Tighter fiscal policy and disruptions from labor unrest weighed on activity amid a backdrop of softening U.S. manufacturing demand and persistent declines in petroleum production. Recent indicators suggest some improvement in the second quarter, although uncertainty regarding trade relations with the United States appears to have increased. In Brazil, real GDP also contracted in the first quarter, as a mining disaster and ongoing weakness in the Argentine economy weighed on Brazilian economic activity. Investment continued to decline, held down by uncertainty over whether Brazil's government would enact major fiscal and other economic reforms.
Financial conditions in many emerging market economies improved, on net, despite the reemergence of trade tensions
Financial conditions in many emerging market economies (EMEs) eased earlier in the year in response to the more accommodative policy stance of the Federal Reserve and major AFE central banks. However, in recent months, political uncertainties in some EMEs and renewed trade tensions between the United States and major trading partners have weighed on EME asset prices. On net, broad measures of EME sovereign bond spreads over U.S. Treasury rates are down a little, while benchmark EME equity indexes are a bit higher since the beginning of the year. Flows to dedicated EME mutual funds increased earlier in the year but turned negative in the second quarter (figure 43). While deteriorations in asset prices and capital flows have been sizable for some economies, particularly Turkey and Argentina, broad indicators of financial stress in EMEs are below those seen during other periods of significant stress in recent years.
The dollar depreciated a little
Over the first half of the year, the foreign exchange value of the U.S. dollar fluctuated but was, on net, a little lower (figure 44). Increased investor optimism about prospects for trade negotiations early this year as well as downward-revised expectations for U.S. interest rates led to a depreciation of the dollar. But the more accommodative tone of communications from major foreign central banks and safe-haven flows—in part in response to trade tensions and concerns about global growth—helped push the dollar up. In addition, the Chinese renminbi has come under some downward pressure since trade tensions escalated in recent months.
1. Owing to population growth, roughly 115,000 to 145,000 jobs per month need to be created, on average, to keep the unemployment rate constant with an unchangedlabor force participation rate. However, the participation rate fell over the December to May period, reducing the number of job gains that would have been needed. There is considerable uncertainty around these estimates, as the difference between monthly payroll gains and employment changes from the Current Population Survey (the source of the unemployment and participation rates) can be quite volatile over short periods. Return to text
2. See the most recent economic projections that were released after the June FOMC meeting in Part 3 of this report. Return to text
3. The Atlanta Fed's measure differs from others in that it measures the wage growth only of workers who were employed both in the current survey month and 12 months earlier. Return to text
4. In the first quarter, labor productivity surged 3-1/2 percent at an annual rate, bringing the four-quarter change to 2-1/2 percent, reflecting a strong pickup in business-sector output and unusual weakness in hours relative to measured gains in payroll employment. This weakness is attributable to a steep decline in a volatile component of hours that is not directly measured in the Bureau of Labor Statistics' establishment survey. Return to text
5. The increases in tariffs on imported goods last year likely provided only a small boost to inflation in 2018 and in the first half of this year. Return to text
6. The trimmed mean index excludes whichever prices showed the largest increases or decreases in a given month. Note that, since 1995, changes in the trimmed mean index have averaged about 0.3 percentage point above core PCE inflation and 0.2 percentage point above total PCE inflation. Return to text
7. Published import price indexes exclude tariffs. However, tariffs add to the prices that purchasers of imports actually pay. Return to text
8. Inflation compensation implied by the TIPS breakeven inflation rate is based on the difference, at comparable maturities, between yields on nominal Treasury securities and yields on TIPS, which are indexed to the total consumer price index (CPI). Inflation swaps are contracts in which one party makes payments of certain fixed nominal amounts in exchange for cash flows that are indexed to cumulative CPI inflation over some horizon. Inflation compensation derived from inflation swaps typically exceeds TIPS-based compensation, but week-to-week movements in the two measures are highly correlated. Return to text
9. As these measures are based on the CPI inflation index, one should probably subtract about 1/4 percentage point—the average differential with PCE inflation and CPI inflation over the past two decades—to infer inflation compensation on a PCE price basis. Return to text
10. It is worth noting that gross domestic income (GDI) has been notably weaker than GDP. GDI is reported to have risen only 1.7 percent in the first quarter relative to the same period of a year ago, 1-1/2 percentage points less than measured GDP growth. GDP and GDI measure the same economic concept, and any difference between the two figures reflects measurement error. Return to text
11. Recently, a large aircraft manufacturer slowed its production and temporarily halted deliveries of an aircraft model. This production slowdown lowers manufacturing output and generates a small drag on real GDP growth in the first half of the year. Return to text
12. The SLOOS is available on the Board's website at https://www.federalreserve.gov/data/sloos/sloos.htm. Return to text
13. The Joint Committee on Taxation estimated that the Tax Cuts and Jobs Act would reduce average annual tax revenue by a little more than 1 percent of GDP starting in 2018 and for several years thereafter. This revenue estimate does not account for the potential macroeconomic effects of the legislation. Return to text
14. The results of the Survey of Primary Dealers and the Survey of Market Participants are available on the Federal Reserve Bank of New York's website at https://www.newyorkfed.org/markets/primarydealer_survey_questions.html and https://www.newyorkfed.org/markets/survey_market_participants, respectively. Return to text