Written by Steven Hansen
Econintersect‘s Economic Index forecast improved this month, but most of the data which forms the basis of our index has yet to capture any impact of the coronavirus pandemic. There is some data which is giving a slight indication but none of the data is showing anything significant. However, it does not take an economist to know that the economy has likely entered a recession. This forecast includes our best guess of the economic impact of the coronavirus pandemic.
Please share this article – Go to the very top of the page, right-hand side for social media icons.
Analyst Summary of this Economic Forecast
This is a black swan economic event. Black swan events are unexpected at the time with immediate significant impact. The speed which the coronavirus engulfed the economy was swift. Economic forecasting tools are not designed to anticipate black swans. We are not alone in publishing an economic forecast which is off the mark – except for a few cases such as ECRI’s Weekly Leading Index – all forecasts suffer the same malady. And even in the case of ECRI’s WLI – it is projecting the decline 6 months from today.
Liberty Street Economics released a non-public version of their weekly economic indicator which supports ECRI’s significant decline – however, this decline was almost entirely due to surging unemployment claims whilst ECRI’s is based on financial data:
Economists are well-practiced at assessing real activity based on familiar aggregate time series, like the unemployment rate, industrial production, or GDP growth. However, these series represent monthly or quarterly averages of economic conditions, and are only available at a considerable lag, after the month or quarter ends. When the economy hits sudden headwinds, like the COVID-19 pandemic, conditions can evolve rapidly. How can we monitor the high-frequency evolution of the economy in “real time”?
To address this challenge, we compute a Weekly Economic Index (WEI) to measure real economic activity at a weekly frequency. Few of the government agency data releases macroeconomists often work with are available at weekly or higher frequency. While financial data, like stock market prices and interest rates, are available at high frequency, we are particularly interested in real activity, not financial conditions. For our purpose, most weekly series come from private sources like industry groups, which collect data for the use of their members, or from commercial polling companies.
The table below details the series we use in our baseline index. These include a measure of same-store retail sales, an index of consumer sentiment, initial unemployment insurance (UI) claims, an index of temporary and contract employment, a measure of steel production, a measure of fuel sales, and a measure of electricity consumption. We transform all series to represent 52-week percentage changes, which also eliminates most seasonality in the data. As the current situation evolves, we may incorporate additional series to refine the index in the coming weeks.
The Atlanta Fed noted on their GDP forecasting web page:
In particular, it does not capture the impact of COVID-19 beyond its impact on GDP source data and relevant economic reports that have already been released. It does not anticipate the impact of COVID-19 on forthcoming economic reports beyond the standard internal dynamics of the model.
Note our comment in March’s economic forecast:
Looking forward, we are concerned about the economic knock-on effects of the coronavirus in China. Simply, Chinese components are widely used across the goods production sector in the U.S. and most other countries in the world. One smart person said, “you cannot ship goods which are not 100 % complete”. If the coronavirus’ effects abates quickly, the disruption will be minor. If it goes on for longer than a few months, a goods shortage will develop in the U.S.
It is interesting that the supply chain issues I was worried about ended up not manifesting to any significant degree as the country and the economy hit the proverbial fan before the supply issues could manifest. The U.S. Secretary of Health and Human Services (HHS) declared a public health emergency on January 31, 2020, with President Trump declaring a national emergency on March 13, 2020. Unfortunately, the attention of politicians was focused on impeaching President Trump (the Senate impeachment vote was on February 5, 2020) – and the swing in focus to the coronavirus was considered a staged political event to change the national dialog away from impeachment.
And even the yield curve is not inverted at this time.
So why even continue reading as this forecast? This forecast is documenting the economy as it would have existed had the coronavirus never visited the economy. This forecast continues to show a weak but improving economy. Our thoughts on what to expect economically in this black swan event:
Everyone is guessing on the way this coronavirus pandemic will play out. The best guesses change daily. The current range of guesses is that the worst will be over in 30 days – to this pandemic continuing through next winter. Because the U.S. is not collecting data in a scientific way [which would require at least a testing of sample population groups for coronavirus], the real rate of spread is unknown. It will be difficult to understand when the worst of the pandemic will be over – so I would guess in the best case the economy could begin to fire up 60 days from today.
This crisis has three sides economically
The shelter in place portion which likely would reduce 2020 estimated GDP by 0.5 % per month for each month the shelter in place portion remains. Without the pandemic, annual GDP would have been in a range of 2.0 % to 2.5 %. So a two month shelter in place would reduce 2020 GDP to 1.0 % to 1.5 %. And I would assume that the shelter in place would not end cleanly in all areas – it will linger in major population centers.
The knock-on effect is more exponential. The longer this crisis continues, the less likely the economy can return to its current trend line. Just like what happened in the Great Recession, the economy would reset. Many businesses would no longer exist – and there is the associated unemployment. It is hard to restart a wounded economy.
There is a high possibility of a knock-on affect triggering defaults in municipalities, states, corporate or private debt.
The “new normal” reset of the Great Recession has been destroyed by the cornavirus. There will be a “post-pandemic normal” reset – and we will have to wait to see what it will be!
At this point, I would project that even in the best case the U.S. will have no economic growth in 2020 – A recession likely began in March 2020. I do not want to speculate on the worst case as there are too many variables – but economically it is ugly.
The rest of the narrative in this economic forecast is based on the current published data – and shows what the economy would have looked like sans pandemic.
Our index’s design is to forecast Main Street growth, whilst GDP is not designed to focus on the economy at Main Street level. One can suggest that GDP is a lagging indicator of the underlying economy – and does not accurately portray the strength and trends of the Main Street economy.
It is interesting that currently, the New York Fed’s Nowcast suggests 2Q2020 GDP of 0.3 % whilst for the 1Q2019 the Atlanta Fed’s GDPNow suggests 2.7 %.
We are concerned about rail transport’s negative growth since the beginning of 2019 – a usual flag for a slowing economy. [click here to see the latest post on rail transport] [click here to see the latest post on truck shipments]
One positive indicator for the Main Street economy is that new home sales continue to have the best growth since 2007. Even existing home sales are now on an improving trend line. [click here to read my take on this subject] [click here to read the latest post on new home sales] [click here for the latest post on existing home sales]
Note that the quantitative analysis which builds our model of the economy does not include housing, personal income, or expenditures data sets.
Econintersect checks its forecast using several alternate monetary-based methods – and all indicate a slow-growth economy – but some are trending up, others trending down, and one in contraction.
Our employment forecast is forecasting POOR employment growth but is on an improving trend.
Note that the majority of the graphics auto-update. The words are fixed on the day of publishing, and therefore you might note a conflict between the words and the graphs due to new data and/or backward data revisions.
The graph below plots GDP (which has a bias to the average – not median – sectors) against the Econintersect Economic Index (which has a bias to median).
This post will summarize the:
- special indicators,
- leading indicators,
- predictive portions of coincident indicators,
- review of the technical recession indicators, and
- interpretation of our own index – Econintersect Economic Index (EEI) – which is built of mostly non-monetary “things” that have been shown to be indicative of the direction of the Main Street economy at least 30 days in advance.
- our six-month employment forecast.
Special Indicators:
The consumer is not spending all of their income – the ratio between spending and income is below the average of the levels seen since the Great Recession.
Seasonally Adjusted Spending’s Ratio to Income (an increasing ratio means Consumer is spending more of Income)
The St. Louis Fed produces a Smoothed U.S. Recession Probabilities Chart which is currently giving no indication of an oncoming recession.
Smoothed recession probabilities for the United States are obtained from a dynamic-factor markov-switching model applied to four monthly coincident variables: non-farm payroll employment, the index of industrial production, real personal income excluding transfer payments, and real manufacturing and trade sales. This model was originally developed in Chauvet, M., “An Economic Characterization of Business Cycle Dynamics with Factor Structure and Regime Switching,” International Economic Review, 1998, 39, 969-996. (http://faculty.ucr.edu/~chauvet/ier.pdf)
Sahm Recession Indicator signals the start of a recession when the three-month moving average of the national unemployment rate (U3) rises by 0.50 percentage points or more relative to its low during the previous 12 months.
Econintersect reviews the relationship between the year-over-year growth rate of non-farm private employment and the year-over-year real growth rate of retail sales. This index has returned to positive territory. When retail sales grow faster than the rate of employment gains (above zero on the below graph) – a recession is not imminent. However, this index has many false alarms.
Growth Relationship Between Retail Sales and Non-Farm Private Employment – Above zero suggests economic expansion
GDPNow
The growth rate of real gross domestic product (GDP) is the headline view of economic activity, but the official estimate is released with a delay. Atlanta’s Fed GDPNow forecasting model provides a “nowcast” of the official estimate prior to its release. Econintersect does not believe GDP is a good tool to view what is happening at Main Street level – but there are correlations.
Latest estimate: 2.7 percent – March 27, 2020
The GDPNow model estimate for real GDP growth (seasonally adjusted annual rate) in the first quarter of 2020 is 2.7 percent on March 27, down from 3.1 percent on March 25. After this morning’s and yesterday’s data releases from the U.S. Census Bureau and the Bureau of Economic Analysis, the nowcasts of first-quarter real gross private domestic investment growth and first-quarter real personal consumption expenditures growth decreased from 7.4 percent and 2.2 percent, respectively, to 4.5 percent and 1.6 percent, respectively, while the nowcast of the contribution of change in real net exports to first-quarter real GDP growth increased from 0.02 percent to 0.45 percent
z forecast8.png or source
Nowcast
The New York Fed also has introduced its own economic projection called Nowcast. Its current forecast:
Mar 27, 2020: New York Fed Staff Nowcast
- The New York Fed Staff Nowcast stands at 1.7% for 2020:Q1 and 0.3% for 2020:Q2.
- News from this week’s releases, covering data for February and 2019:Q4, increased the nowcast for 2020:Q1 by 0.2 percentage point and increased the nowcast for 2020:Q2 by 0.2 percentage point.
- Positive surprises from February’s manufacturing data accounted for most of the increase.
z forecast7.png
A yield curve inversion historically has been an accurate predictor of an impending recession. A yield curve inversion is where short term bonds have a higher yield than longer-term bonds. The graph below shows inversions prior to USA recessions. Econintersect does not believe the yield curve is a reliable indicator of recessions in the New Normal where monetary policy uses extraordinary tools.
Special Indicators Conclusion:
Most economic releases are based on seasonally adjusted data which are revised for months after issuance. The real trends in a particular release may not be obvious for many months due to data gathering and seasonality adjusting methodologies. The special indicators are showing slow economic growth, but a marginal indication of a recession is in the inverted yield curve which we do not believe is an accurate indicator.
The Leading Indicators:
The leading indicators are for the most part monetary based. Econintersect‘s primary worry in using monetary-based methodologies to forecast the economy is the accommodative monetary policy which may (or may not) be affecting historical relationships.
Econintersect does not use data from any of the leading indicators in its economic index. Leading indices in this post look ahead six months – and are all subject to backward revision.
Chemical Activity Barometer (CAB) – The CAB is an exception to the other leading indices as it leads the economy by two to fourteen months, with an average lead of eight months. The CAB is a composite index that comprises indicators drawn from a range of chemicals and sectors. It is a relatively new index and appears somewhat accurate (but its real-time performance is unknown – you can read more here). A value above zero is suggesting the economy is expanding.
ECRI’s Weekly Leading Index (WLI) – A positive number shows an expansion of the business economy, while a negative number shows contraction. The methodology used in creating this index is not released but is widely believed to be monetary based.
Current ECRI WLI Index
The Conference Board’s Leading Economic Indicator (LEI) – the LEI has historically begun contracting well before a recession but has had many false contractions.
z conference1.png
Leading Index for the United States from the Philadelphia Fed – This index is the super index for all the state indices. This index can have a significant backward revision and is considered close to worthless because of this backward revision.
Nonfinancial leverage subindex of the National Financial Conditions Index – a weekly index produced by the Chicago Fed signals both the onset and duration of financial crises and their accompanying recessions. Econintersect now believes this index may be worthless in real-time as the amount of backward revision is excessive – so we present this index for information only. This index was designed to forecast the economy for six months in advance. The chart below shows the current index values, and a recession usually occurs months to years after the trend line changes from positive to negative.
Leading Indicators Conclusion: The takeaway is a soft but generally improving economy.
- Chemical Activity Barometer (CAB) growth rate is improving and now is in expansion – but still indicating a soft economy.
- ECRI’s WLI is now forecasting a recession.
- The Conference Board (LEI) 6 month rolling average suggesting a slow but improving growth rate over the next 6 months. Some have suggested that it is indicating a recession – but this index has dropped below zero before without a recession ensuing.
- The Philly Fed’s Leading Index continued backward revisions make this index worthless – however its growth trend currently is at the average of the values seen in the last two years.
Forward-Looking Coincident and Lagging Indicators
Here is a run-through of the most economically predictive coincident indices which Econintersect believes can give up to a six-month warning of an impending recession – and do not have a history of producing false warnings. Econintersect does not use any of these indicators in its economic forecast.
Consider that every recession has different characteristics and dynamics – and a particular index may not contract during a recession, or start contracting after the recession is already underway.
Truck transport portion of employment – to search for impending recessions. Look at the year-over-year zero growth line. For the last two recessions, it has offered a six-month warning of an impending recession with only one false warning. Transport is an economic warning indicator because it moves goods well before final retail sales occur. Until people stop eating or buying goods, transport will remain one of the primary economic pulse points. When this sector turns robotic in the coming years – this measure will become useless – but currently, the shift from box stores to e-Commerce is creating much more employment in this sector. Either way – this index may not be capable of alerting the next recession.
Transport employment growth is at the zero growth line. As transport provides a six-month recession warning – the implication is that any possible recession is six months away.
Business Activity Markit US Services Activity Index – this index is noisy. The index is now below 50 (below 55 is a warning that a recession might occur, whilst below 50 is almost proof a recession is underway). This index may not provide timely warnings of recessions.
z%20markit_services.png
US Treasury Tax Receipts – For the Great Recession, the rolling averages went negative in February 2008 – two months after the Great Recession’s start. For the 2001 recession, the rolling averages for tax revenues went negative two months after the official start of the recession. Currently, the year-over-year rolling average growth is up 8.7 % year-over-year – up from last month’s +4.5 %.
Year-over-Year Change in US Government Receipts – Monthly (blue line) and Three Month Rolling Average (red line)
z forecast5.png
Predictive Coincident Index Conclusion: The predictive indices are mixed relative to economic trends – and one is suggesting a recession is underway.
Technical Requirements of a Recession
Sticking to the current technical recession criteria used by the NBER:
A recession is a significant decline in economic activity spread across the economy, lasting more than a few months, normally visible in real GDP, real income, employment, industrial production, and wholesale-retail sales. A recession begins just after the economy reaches a peak of activity and ends as the economy reaches its trough. Between trough and peak, the economy is in an expansion. Expansion is the normal state of the economy; most recessions are brief and they have been rare in recent decades.
… The committee places particular emphasis on two monthly measures of activity across the entire economy: (1) personal income less transfer payments, in real terms and (2) employment. In addition, we refer to two indicators with coverage primarily of manufacturing and goods: (3) industrial production and (4) the volume of sales of the manufacturing and wholesale-retail sectors adjusted for price changes.
Below is a graph looking at the month-over-month change (note that multipliers have been used to make changes more obvious).
Month-over-Month Growth Personal Income minus transfer payments (blue line), Employment (red line), Industrial Production (green line), Business Sales (orange line)
In the above graph, if a line falls below the 0 (black line) – that sector is contracting from the previous month. Only industrial production growth remains in negative territory. Another way to look at the same data sets is in the graph below which uses indexed real values from the trough of the Great Recession.
Indexed Growth Personal Income minus transfer payments (red line), Employment (green line), Industrial Production (blue line), Business Sales (orange line)
NBER Recession Marker Bottom Line – this month three of the four recession markers are in positive territory.
Econintersect believes that the New Normal economy has different dynamics than most economic models are using.
Economic Forecast Data
The Econintersect Economic Index (EEI) is designed to spot Main Street and business economic turning points. This forecast is based on the index’s three-month moving average. The three-month rolling index value is now a positive 0.148 – an improvement from last month’s positive 0.027. The economic forecast is based on the 3-month moving average as the monthly index is very noisy. A positive value of the index represents Main Street’s economic expansion. Readings below 0.4 indicate a weak economy, while readings below 0.0 indicate contraction.
A summary of elements affecting our economic index:
- The government portion relating to business and Main Street are little changed.
- The business portion is slow but the improvement in our economic index comes mostly from the business sector.
- The consumer portion rate of growth is weak but remains little changed.
The EEI is a non-monetary based economic index that counts “things” that have shown to be indicative of the direction of the Main Street economy. Note that the Econintersect Economic Index is not constructed to mimic GDP (although there are correlations, the turning points may be different), and tries to model the economic rate of change seen by business and Main Street. The vast majority of the inputs to this index uses data not subject to backward revision.
The red line on the EEI is the 3 month moving average.
Consumer and business behavior (which is the basis of the EEI) either lead or follow old fashion industrial age measures such as GDP depending on the primary dynamic(s) driving the economy. The Main Street sector of the economy lagged GDP in entering and exiting the 2007 Great Recession.
As Econintersect continues to backcheck its model, from time-to-time slight adjustments are made to the data sets and methodology to align it with the actual coincident data. To date, when any realignment was done, there have been no changes for trend lines or recession indications. Most changes to date were to remove data sets that had unacceptable backward revisions, became too volatile, or were discontinued. The last realignment was done in the June 2016 forecast to swap an industrial production data set which became too volatile. Documentation for this index was in the October 2011 forecast.
Jobs Growth Forecast
The Econintersect Jobs Index is forecasting non-farm private jobs growth of 80,000 for March – higher than last month’s forecast of 70,000.
Summary of Jobs Forecast
The fundamentals which lead job growth are now showing an improving growth trend in the employment growth dynamics. We are currently predicting the jobs growth six months from today will exceed the growth needed to maintain participation rates and the employment-population ratios at the current levels.
The Econintersect Jobs Index is based on economic elements that create jobs, and (explanation here) measures the historical dynamics which lead to the creation of jobs. It measures general factors, but it is not precise (quantitatively) as many specific factors influence the exact timing of hiring. This index should be thought of as a measurement of job creation pressures.
For the last year, job growth year-over-year (green line in below graph) is bouncing around. The forecast by the Econintersect Jobs Index is shown as the blue line in the below graph. A fudged forecast (red line in below graph) is based on the deviation between forecast & current actual using a 3-month rolling average.
The fudge factor (based on the deviation between the BLS actual growth and the Econintersect Employment Index over the last 3 months) would project jobs growth at 190,000.
Analysis of Economic Indicators:
Econintersect analyzes all major economic indicators. The table below contains hyperlinks to posts. The right column “Predictive” means this particular indicator has a leading component (usually other than the index itself) – in other words, it has a good correlation to future economic conditions.
General Economic Indicators:
Monthly Data: {click here to view full screen}
Quarterly Data: {click here to view full screen}
Aruoba-Diebold-Scotti Business Conditions Index: {click here to view full screen}
include(“/home/aleta/public_html/files/ad_openx.htm”); ?>
