This is not an easy task. This is actually an extremely tough process, which demands constant analysis and appropriate interpretation of hundreds of different macro statistics, trends, incoming data and events. Big banks have massive departments filled with statistitians, mathematicians, macroeconomic advisors and all other possible wise guys who do nothing else all days long, but put together enormus amounts of data in an effort to answer one simple but extremely important question: how is the economy doing? Answering this question is of huge importance to all of us – employees, employers, moms and dads and kids, producers and consumers, exporters and importers, businessmen, politicians, the young and the old, manufacturers and service providers, bankers, brokers, the rich and the poor and last (but not least) the investors. Short – it’s the clue question to everyone. The answer to this question determines our lives in every possible material sense. If the economy is doing OK, it means healthy labour market (work for all of us), calm political situation, enough taxes for the country’s fiscal spending plans, stable everyday business circumstances, money earned, stable growth of aggregate income, social rest and last (but again not least) stable markets, which can be forecasted, modelled and predicted. If the economy is falling off the cliff though, the situation is obviously totally opposite. There’s lost jobs, lost income, unrest, instability, unpredictability, lack of trust, losses, inability to forecast, modell and predict and a lot of volatility. Noone wants that and everyone is afraid of it. Also capital markets are obsessed with recession fears. Hence it is crutial for an investor to understand at which point of the economic cycle the global economy is at any given point to accordingly adjust their behaviour on the markets. An investor’s portfolio should look totally different in a recession and in a recovery. You don’t want to be 90% exposed to equities as recession comes, you want to here close to zero equities then! Otherwise you’ll possibly have to wait another decade for you equities holdings to recover their losses fully and this can be a painful long lesson.
Now we’re coming to a point where we need to get to know whether forecasting economic cycles can be simplified to a degree that can be comprehended by one person who is not necessarily an economic specialist, but wants to manage his or her money wisely to minimize the risk of big losses. In my view the answer is YES. If you ask yourself what are the most crutial macroeconomic indicators showing the biggest possible spectrum of current and expected conditions in the economy, you possibly come to this answer: there is 3 major readings: 1. Inflation (a reading showing how fast the prices change on aggregate basis – to fast a growth is not good, but a fall is not telling a good story either) 2. Unemployment rate (an index telling us whether the labour markets are in good shape or maybe in trouble as people lose jobs with all the negative consequences for their future lives) 3. Interest Rates (or put simply – the price of money). In short term the price of money is set by the Central Bank (CB), an institution responsible for money creation (Short Term Interest Rates). In the longer run though other factors also influence the price of money that is out of CB’s control (it’s mostly time and inflation expectations) (Long Term Interest Rates).
The above factors have their natural ways of behaviour accordingly to the phase of the cycle and some natural relations between each other. Economy is always following a sinusoidal pattern – the ups and downs coming after another and repeat constantly. The economic cycle can very generally be split into 4 major phases: Recovery, Expansion, Boom, Recession.
In the Recovery phase usual patterns are:
(a) unemployment is still high after Recession, but shows tendency to stabilize at high level or even fall slowly
(b) inflation is usually low and falling or reaching lows
(c) Short Term Interest Rates (STIR) are low as CB is OK to keep stimulating the economy after tough Recession times
(d) Long Term Interest Rates (LTIR) are usually not very high, but higher than STIR (which is a natural and healthy relation betweeen STIR and LTIR)
(e) output starts to rise in the economy.
In the Expansion phase usual patterns are:
(a) unemployment is steadily falling as the economic situation improves, people find jobs, there is more and more work available
(b) inflation is off its lows but moderate without signs of sudden acceleration
(c) STIR are lower than LTIR but CB is trying to adjust them higher as economy evolves to withdraw the unnecessary stimulus (the economy can usually run on its own mostly)
(d) LTIR are rising as higher inflation expectations amongst investors and participants of economic life are anchoring into their minds and behaviour
(e) output is rising
In the Boom phase usual patterns are:
(a) unemployment is low and reaching its lows, the labour market situation is usually very good, jobs are plenty and people can get hired easily, but there are 1st signs of instability as some employers start feeling unsafe about their business prospects
(b) inflation is usually accelerating towards levels higher than during Expansion phase
(c) STIR are rising as CB is limiting access to money in efforts to fight inflation rising above acceptable levels (ate around peak of cycle CB is fully restrictive in its policy)
(d) LTIR are usually falling towards (or at least growing slower than) STIR as agressive CB swings market’s expecations of future inflation lower. This phenomenon is called The Yield Curve Flattening or Inversion – this is not a heatlhy situation, it usually means troubles for economy in forseeable future (inversion of the yield curve is by itself not a sign Recession is immediate)
(e) there are rising output constraints in the economy and many assets are inflated by historical standards.
In the Recession phase usual patterns are:
(a) unemployment is rapidly rising, people lose jobs and have great difficulties finding new ones, employers are very uncertain about their future and firing and cutting job postings is normal
(b) inflation is falling off the cliff as people make sudden cuts in spending due to worsening employment prospects (falling consumer inflation or CPI) and companies face substantial worsening of their financial standing and struggle to sell their products or services (falling producer inflation or PPI)
(c) STIR are falling rapidly as CB is making all possible efforts to make money cheap to stimulate the economy
(d) LTIR are falling rapidly as peoples’ expectations about future prices dynamics in the economy are heavily dumped (there’s usually even plenty of fears of deflation, so falling prices, which are extremely destructie to the economy as they discourage consumption causing the so called it-will-be-cheaper tomorrow effect); LTIR are below STIR at the start of Recession usually, but then tend to rise as CB keeps cutting STIR agressively, causing inflation expectations to rise eventually – the yield curve slowly comes back to its usual shape where LTIR>STIR as the economy heads towards Recovery again
(e) output is falling sharply.
And so it goes round and round…
Now what’s the use of it all from an investor’s perspective? Economic cycles determine not only real economy, but also the pricing of all kinds of assets, whether it’s real estate, land, bonds, equities or currencies. All these assets are usually valued much differently in Recovery, Expansion or Boom or Recession. An investor must be able to determine the point in economic cycle before deciding on his/her portfolio structure sensibly. Not having an army of analysts behind you is not necessarily making it that easy, as I mentioned above. But what if we tried to simplify the whole process and pack as much of the macro data into a one relatively simple, but very comprehensive index or method, which would maximally increase our chances of getting the right answer to the most important question we want to answer: where are we in the economic cycle?
In my personal view the absolutely most valuable method forecasting economic cycles I have ever run into is Doctor Robert Dieli’s Enhanced Aggregate Spread (EAS) called Mr.Model. It fulfills all the requirements of a simple but comprehensive method I have long been looking for. Additionally it’s backtested forecating power is astonishing. And the cherry on the cake, the way Dr.Dieli put it together and explains it to his supporters (including myself) is so simple that even people with no economic background could easily understand it. A perfect method?! Let’s take a look at Mr.Model. Again, all credits to Doctor Robert Dieli from NoSpinFoceast (https://www.nospinforecast.com).
Mr.Model aims to 1st determine in which phase of the cycle the economy is at certain point in time and 2nd figure out when and why the economy will move into the next phase of the cycle. Mr.Model describes the US economy (not global economy, but as I wrote above, think the US economy is the best and most important one to follow for somone who runs his portfolio with more or less global exposure). It’s obviously not Mr.Model that sets the dating of the economic cycles. It is The National Bureau of Economic Research [NBER], based in Cambridge, Massachusetts. The Business Cycle Dating Committee of the NBER is the body that sets official dates of when the US economy is in expansion or contraction. Point is that once you get to know it from NBER, the economy has already trespassed the phase-change point (read: the assets allocation of the portfolio you’re running should have been adjusted already in the past; if you still have 80% of stocks in your portfolio a few months into Recession, it’s too late, possibly 30-40% of the portfolio’s value is already gone). The trick with forecasting models is that they try to give us and answer of cycle phase change before it really happens. Mr.Modell has a 9-12 month forecasting period. How is this possible? It’s because the variables it usues and the relations between them have the forward-looking bias. More on this later.
EAS consists of two separate spreads. One of them, the EFS (Enhanced Financial Spread), monitors the financial conditions in the economy via the yield curve (in other words: the different between the long term government debt and the Central Bank’s (Fed’s) shortest possible price of money) . EFS formula is:
EFS = 10-year US Treasury Bonds yield [10yr] minus Fed Funds Rate [FFR].
Why such a spread and what does it tell us? The 10yr gives us a grip on time value of money in the future, inflation risk (pretty important from an investor’s point of view), market sentiment (investors use 10yr treasuries as world’s major proxy of save haven and the risk of a default in case of US government is close to zero; in case of turmoil in capital markets, investors „hide” in US bonds, not to just sit on cash). The sentiment-catching characteristics of 10yr, especially in relative terms vs. FFR, is very important as it gives us indications on what the market thinks about future developments in the economy. If the yield curve shape is normal (10yr is higher than FFR) is tells us that: value of money is positive as it normally should be, investors accept only higher yields for holding longer-dated treasuries, investors perceive inflation risk as positive (whilst recession usually means deflation or negative inflation risk), investor sentiment is generally positive (they do not „hide” in treasuries as they expect the economy to grow in the future). On the other hand a flat or negative yield curve (10yr at or below FFR) is telling a totally opposite story: value of money is zero or negative (not normal right?), investors expect low or negative inflation (deflation), the market sentiment is negative as investors buy treasuries just for the sake of not being in stocks (or other risky assets) or cash (zero-yield asset), added all up – people are fearful of troubles in the economy ahead. The other component of the EFS, the FFR, is basically the instrument the Fed controls to adjust the price of shortest-term transactions with the banks. It is the measure of Fed’s attitude to fulfilling its two major monetary goals – keeping inflation under control and keeping employment as high as possible. The Fed can be agressive or hawkish (and raise FFR above its natural level seen as neutral to the economy) in trials to either fight inflation or certain asset bubbles or just slow down the booming economy. the Fed can also be very accomodative or dovish (and cut FFR below natural level) to support an ailing economy. The Fed can also be just neutral and let the the economy grow at it’s natural rate, assumed the FOMC (Fed Open Market Committee – the Fed’s deciding body) sees no threats of instability in prices or labour market.
The second spread, ERS (Enhanced Real Spread), measures the tendencies in real economy by including CPI (Consumer Price Index – most widely used inflation measure) and STRUC (Short Term Unemployment Rate, measured by diving the number of people out of work for less than 27 weeks by total civilian workforce). Here’s the ERS formula:
ERS = CPI minus STRUC.
CPI measures inflation. In normal circumstances CPI is moderate and shows no tendencies to either spike or plunge rapidy. This is a characteristic of a normal economy growing at or around its natural rate. STRUC is usually low and falling in a normal economy. If a number of short-term unemployed starts to rise, it tells us that the labour market is getting worse, employers stop hiring or even start firing expecting slowdown or recession.
We have now arrived at EAS calculation:
EAS = EFS minus ERS
EAS = (10yr – FFR) – (CPI – STRUC)
You possibly subcutaneously start getting the point here right? EAS must be positive in normal conditions of expanding economy, because EFS is postive in such situations (read above if forgotten) and ERS is negative in normal conditions, because CPI is usually lower than STRUC (Fed has a 2% CPI target and doesn’t like it much higher for long, plus STRUC below say 3.5-3% is only frictional unemployment of those who are just in the process of changing jobs). A positive ERS is not something we like to see, it usually means inflation running out of Fed’s control – one should then expect the Fed’s appropriate (hawkish) reaction and a possible cycle peak ahead.
The situation we’re looking for while trying to determine the next cycyle peak (and appropriately change the allocation of our portfolio before it actually arrives, not after) is when the positive EFS goes down through zero and negative ERS goes up through zero. a moment when they cross each other around the point zero, whilst going in different directions (as stated above). Such a situation is packed with alarming signals: an inverting yield curve (see above on what it means for investors’ minds), agressive Fed having to fight growing inflation that’s out of its long-term targets. A cycle peak is imminent some time in the future. As backtesting of the model done by Dr. Dieli showed, a 9 months leading period was the most accurate. In only 2 cases since 1955 the EAS model has given false signals about upcoming cycle peaks with 9 months lead. Both of them were in 1960’s. One time it was a false negative signal (a recession in 1960-61 whilst EAS was around 300 pts) and the other one was a false positive (no recession with EAS having dipped below 0 pts in 1967). Said short, EAS correctly predicted 8 out of 9 recessions since 1955 and once it predicted a recession that didn’t come. Depending on how you interpret it, that’s either 90% or 80% success rate. That is something right?
Now moving on. A 9 months lead is obviously quite a bit of time to appropriately prepare the structure of a portfolio somone runs for worse times. It also is a time when an investor shall be much more sensitive about what’s happenning to the macro data, to see if they do start confirming the EAS’s forecast during the 9 months between the model’s cycle peak and the actual peak at the time of it happening. To help strenghten the model, Dr.Dieli incorporated 3 so called coincident indicators into the forecasting procedures:
(a) Private Payrolls (YoY change)
(b) Industrial Production (YoY change)
(c) DeltaDelta (measured as Percent Change from Prior Year Month of the S&P500 Stock Index minus the Percent Change from the Prior Year Month of the Official Unemployment Rate).
Why those three measures? YoY change in Private Payrolls shows the dynamics in the labour market. YoY change in Industrial Production shows the dynamincs in Output. DeltaDelta shows very nicely the relative dynamics between the stock market and unemployment (positive DeltaDelta means growing stocks and stable or falling unemployment, a negative DeltaDelta tells us that the stock market dynamic over last year starts to predict troubles ahead on the labour market, which are usually equal to recession).
If we got an imminent cycle peak reading from EAS, we want to watch what’s happening with the coincident indicators. Their recession associated levels are as follows: Private Payrolls <1.00% and falling, Industrial Production <1.00% and falling, DeltaDelta <0.0% and falling. The author additionally takes a range of 0-200 pts for the EAS as a so called Danger Zone, which is most exposed to recessions. Indeed history shows that while EAS enters this zone from above, it usually goes down to 0 and below quite fast or at least gets wobbly within that range, sounding out increased probability of a recession due within the forecasting period (note this chart shows EAS+9 so the forecasted value; is effectively known 9 months before it is denoted on this chart at a certain point in time). This can be clearly seen in the case of last 3 recessions on below graph:
Now that we know that once EAS is below 200 pts the economy is recession-eligible, we need to figure out exactly where in the economic cycle we are. To answer that question Mr.Model takes use of the labour market indicator again – the YoY % change of Total Private Nonfarm Payroll Employment (TPNPE) (smoothed out by using a six-month average of that coincident indicator to eliminate short-term moves, which can be blurring the bigger picture). The rules of assessing where the economy is in the business cycle are:
EAS+9 above 200 -> Expansion
EAS+9 below 200 and 6-month moving average of TPNPE:
greater than 2% -> Expansion
falling from 2% to 1% -> Boom
falling from 1% to minimum -> Contraction (Recession)
rising from minimum to 2% -> Recovery.
Simple? I think it is. The below grapf depicts it very clearly:
With all the data above, we can now estimate with high probability where in business cycle we are and when the next turning point might be. With that knowledge an investor can make more educated decisions as to his or her portfolio of risky assets and adjust the risk profile of the portfolio appropriately.
Going forward Mr.Model will be used as a base instrument in aggregate equity allocation decisions in the exemplary portfolios I run in my blog. More on the portfolios in my different posts describing them one by one. Please see the portfolios descriptions here:
Pension Portfolio (Low Risk)
Global Leaders Portfolio (Higher Risk).
I wish you all best of luck!
Disclaimers: None of the ideas, views and thoughts presented here shall ever be taken as a recommendation to buy or sell stocks,bonds,FX,commodities or any other financial instruments as stated in REGULATION (EU) No 596/2014 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 16 April 2014 on market abuse (market abuse regulation) and repealing Directive 2003/6/EC of the European Parliament and of the Council and Commission Directives 2003/124/EC, 2003/125/EC and 2004/72/EC or the Regulation of the Polish Minister of Finance of 19 October 2005 on information constituting recommendations regarding financial instruments, their issuers or exhibitors (Journal of Laws of 2005, No. 206, item 1715) or the Polish Act of 10 February 2017 amending the act on trading in financial instruments and some other acts. The article is for educational reasons and purely presents private views of the author, thus the author shall not be claimed eligibile for any losses of a third party resulting from trading activities based upon this article. The author uses his best knowledge and data from sources believed to be reliable, but makes no representations as to the accuracy of the data. Full Disclaimers&Liability Limitations page.