Apr. 19 2019 — It isn't so doom and gloom after all. The excessive concerns over weakness in first-quarter growth are fading, as evidenced by broad-based strength in retail sales in March, which bounced back decisively, by 1.6%. With limited revisions to the first two months of the quarter, this is still not enough to get real consumer spending very far above 1% for the quarter (the weakest since a year ago). But still, it points to firmer momentum at the end of the first quarter.
Consumer sentiment has also turned up following the end of the government shutdown, and financial conditions have improved quickly over the last three months. Meanwhile, jobless claims fell to new historical lows in April, to start the second quarter, after going through a temporary upturn in January and February. Supplementing the hard data, anecdotal evidence from the Federal Reserve's latest Beige Book also depicts strengthening in economic momentum heading into spring. All combined, they support our view that the weaker early first quarter was a temporary soft patch and that consumer spending growth should get back to around a 3% rate in the second quarter.
First-Quarter Growth Is Running Ahead Of Forecast
Incoming economic data over the past two weeks point to faster-than-anticipated growth in the first quarter, likely closer to 2% than 1% (our published forecast in March was 0.8%). The Atlanta Fed's GDPNow--our preferred running estimate of real GDP growth based on available data for the latest quarter--currently shows 2.8% annualized growth for the first quarter.
Higher-than-expected contributions to growth from net exports and the pace of private inventory accumulation are largely responsible for our growth estimates running behind. Both of these categories are unlikely to be significant contributors to growth because they are unsustainably high currently. Meanwhile, growth in personal consumption expenditures and fixed nonresidential business investment, to date, has lined up well with our forecasts. Government spending and residential investment are slightly higher than our forecast.
That said, our forecast incorporates about a -0.5 percentage point penalty to first-quarter growth--because of residual seasonality. Residual seasonality is when seasonal patterns remain in data that are reported to have been adjusted for seasonal variation. Over the past couple of decades, first-quarter economic growth has been substantially lower than the growth in other quarters, even after adjusting for typical seasonality (see chart 1).
From 2000 to 2018, average growth for the third and fourth quarters (2.0% and 1.9%, respectively) is about equal to average growth over all quarters (2.0%). During the same period, average growth in the first quarters (1.5%) is below the sample average, and average growth in the second quarters (2.8%) is above the sample average. This suggests a residual seasonality bias in which growth is reduced below trend in first quarters, boosted above trend by about the same amount in second quarters, and is unaffected in the third and fourth quarters. The Bureau of Economic Analysis (BEA) has recognized the phenomenon, and efforts have been made to correct it every so often, but the issue remains. (Economic literature on this issue identifies nonresidential structures, government consumption, and exports components of GDP as the primary culprits driving this bias.)
For what it is worth, this discrepancy matters for monetary policymakers. For example, take the Taylor rule (widely considered the benchmark for monetary policy) that is often invoked in discussions about appropriate interest rate. According to the original Taylor rule (from 1993), if the reported GDP were 1% lower than the actual value due to residual seasonality, the changes in output gap from that could mean lowering rates by 0.5 percentage point.
As such, when the first-quarter GDP numbers come out next week, we will be looking for gross domestic income (GDI), in addition to the gross domestic product (GDP), to get a sense of how the economy fared during the quarter. (GDI measures economic activity in terms of income earned, while GDP measures it in terms of production. The two are theoretically equivalent, but they differ slightly in practice due to differences in source data. The BEA tested both GDP and GDI for residual seasonality and found that while GDP does exhibit residual seasonality, GDI does not.)
Housing Starts Struggle To Find Their Footing
Housing starts were softer than expected in March, declining 0.3% to 1.139 million (annualized rate). Part of the reason for lower-than-expected groundbreaking was the severe flooding in the Midwest. But it was also the South (the biggest region for housing activity) that came in weaker than expected. Permits, which are a forward-looking indicator of starts, were not any better, falling 1.7% over the month to 1.269 million. Permits for single-family homes--another key reading--fell 1.1% in March and are down 5.1% year over year.
The trend is clearly downward, with both starts and permits lower than last spring (see chart 2). That said, one piece of good news (in the March report) was an 11.9% monthly jump in single-family completions, to a 938,000 rate for a new home sales market that needs fresh supply.
While the housing sector continues to face the same challenges as it did last year--mainly supply constraints from higher costs of labor and regulations, as well demand headwinds from affordability issues--we see modestly better prospects for the sector this year. Pressures from housing affordability should ease somewhat as interest rates take a pause and the labor market continues to deliver solid outcomes, including higher wage rates. Low mortgage rates may already be helping purchase applications (see chart 3) and what looks like a bounce back in confidence among homebuilders (see chart 4). In the medium to longer term, an improving demographic boost with pent-up demand should support at least a modest upward shift in housing activity.
Since last year, the pace of growth in real residential investment has fallen behind overall real GDP growth. This follows six consecutive years of real residential investment growth outpacing overall real GDP growth, from 2012 to 2017, as the sector benefited from historically low mortgage rates and strong and sustained job growth. We see residential investment remaining a net neutral contributor to GDP growth in 2019, with housing starts likely falling in the range of 1.25 million-1.29 million units for the year. While the real investment in multifamily structures has fully recovered to its prerecession level, real investment in single-family structures hasn't, indicating room for further growth.
Table 1 |
|Review Of Economic Indicators Released In The Past Two Weeks (April 8, 2019-April 19, 2019)|
|Latest period||Apr-19||Mar-19||Feb-19||Jan-19||Level year ago||% year-over-year|
|Jobless claims (April 18, 2019)||April||201,250||218,650||227,125||222,500||223,750|
|Consumer spending and confidence|
|Consumer Sentiment Index (University of Michigan)||April||96.9||98.4||93.8||91.2||98.80|
|Business activity and sentiment|
|Industrial production (m/m, % change)||March||(0.1)||0.10||(0.30)||2.8|
|Capacity utilization (level, rate)||March||78.8||79.0||79.1||78.2|
|Retail sales (m/m, % change)||March||1.6||(0.2)||0.8||3.6|
|Retail sales less auto (m/m, % change)||March||1.2||(0.2)||1.4||3.6|
|Wholesale trade, inventories (m/m, % change)||February||0.3||0.9||4.9|
|Philadelphia Fed General Business Conditions Index||April||8.5||13.7||(4.1)||17.0||23.4|
|Empire State General Business Conditions Index||April||10.1||3.7||8.8||3.9||17.9|
|Factory orders (m/m, % change)||February||(0.5)||0.0||2.4|
|Durable goods orders (m/m, % change)||February||(1.9)||0.3||2.5|
|CB Leading Index (m/m, % change)||March||0.4||0.1||0.0||3.1|
|Import prices (m/m, % change)||March||0.6||1.0||0.1||0.0|
|Export prices (% change)||March||0.7||0.7||(0.6)||0.6|
|Trade balance - goods and services (bil. US$)||February||(49.4)||(51.1)||(55.7)|
|Exports - goods and services (bil. US$)||February||209.7||207.4||204.7|
|Imports - goods and services (bil. US$)||February||259.1||265.300||260.4|
|PPI-final demand (m/m, % change)||March||0.7||0.3||0.1||2.2|
|CPI (m/m, % change)||March||0.4||0.2||0.0||1.9|
|Core CPI (m/m, % change)||March||0.1||0.1||0.2||2.0|
|Note: Jobless claims is four-week average data. Sources: U.S. Bureau of Labor Statistics, U.S. Bureau of Economic Analysis, U.S. Census Bureau, Institute for Supply Management, and ADP Research Institute.|
Table 2 |
|Economic Release Calendar|
|22-Apr||Existing home sales (mil.)||Mar||5.3||5.3||5.51|
|23-Apr||New home sales (mil.)||Mar||0.641||0.648||0.667|
|25-Apr||Durable goods orders (%)||Mar||1.2||0.8||(1.6)|
|26-Apr||GDP advance report (q/q %)||Q1||0.8||1.9||2.2|
|Chain Price Index advance report (%)||Q1||1.8||1.2||1.7|
|University of Michigan Consumer Sentiment (final)||Apr||98||97||96.9|
|29-Apr||Personal income (%)||Mar||0.3||0.4||0.2|
|Personal Consumption Expenditures (%)||Feb||0.3||0.4||0.1|
|30-Apr||Employment Cost Index||Q1||0.8||0.8||0.7|
|1-May||ADP Employment Survey (000s)||Apr||190||178||129|
|Construction spending (%)||Mar||0.2||0.3||1|
|Nonfarm productivity (prelim) (%)||Q1||0.5||0.8||1.9|
|Unit labor costs (prelim) (%)||Q1||2.8||2.5||2|
|2-May||Factory orders (%)||Mar||0.7||0.5||(0.5)|
|3-May||Nonfarm payrolls (000s)||Apr||200||185||196|
|Private nonfarm payrolls (000s)||Apr||185||175||182|
|Manufacturing payrolls (000s)||Apr||11||8||(6)|
|Unemployment rate (%)||Apr||3.7||3.8||3.8|
|Average hourly earnings (%)||Apr||0.2||0.3||0.1|