For a recent analysis of the national outlook see here.

S & P 500


The following chart shows the weekly closing price of the S & P 500 beginning the first Friday in January 2019 thru the current date.  What a blast during the recent weeks.  The long increase in the S & P first underway during Obama’s administration is brought to an end by Trump.  Is any collapse more dramatic than the following?


The end of the decline is unknown in both its level and duration.  Since both the level and duration are unknown, the market faces uncertainty.  Risk is different from uncertainty.  Risk can be hedged and uncertainty cannot.  An example of hedging a risk is home insurance.  The source of risk is a fire. The purchase of fire insurance provides protection (the hedge) against financial loss resulting from fire damage.

Not so with uncertainty because both the future benefits and costs are both unknown.

(NEW) During the past couple of weeks, the market seems to be reacting to (1) the prospects of an anti-virus treatment and (2) reopening the economy.  The reality is (1) an identification, approval, production and distribution of anti-viral treatment is long in the future and (2) opening the economy will prompt more virus outbreaks and opening will stimulate the economy less the many bankruptcies and business closings.

Expect the recovery to be painstakingly slow with constant chatter from Trump attempting to distract the voters from his failures to act. (END)

The forecasts shown below are based on the actual historical patterns of the S & P index plus identifiable historic events (these are named interventions).

If the above chart were extended backward to earlier dates, it would show a significant upward trend with some downward blips (as seen for the short periods May – June 2019, August – September 2019 and mid-Jan – Feb 2020).  An examination of the annual percentage increase in the S & P 500 during the past 5 – 10 years would reveal the increase to be 10 – 12% per compounding period.  Investors following a passive investment strategy, i.e. buying an index fund, would earn a return exceeding the return earned following an active investment strategy with significantly lower risk as measured by the standard deviation of the annual returns.

The following table shows the summary of the actual and forecasts for the weekly change in the S & P 500 for 2020.


The actual change for the year = 81.6.  The discussion in the trade press attributes the blip to the continued virus scare in the U.S. and world, and the President’s slow reaction along with the continuing trade war and the general outlook for the world’s economies . We’ll see during the summer.


The following table shows the summary of the actual and forecasts and actual and forecast for the weekly change in the S & P 500 for 2019.


The sum of the actual forecasts to date for 2019 = 359.72.  The total percentage change from the first to the last Friday in the year = 27.97%.


(NEW) The market increased from the first Friday in January thru the first Friday in April.  Then the market was sideways with a significant increase in the variation (measured by the variance or standard deviation) thru the first week in September.  This side-wise movement reflected the market’s focus on the Trump-China tariff negotiations.  Since the first week of September, the outlook for some resolution of the tariff negotiations increased.  The market resumed its upward climb.

The outlook for 2020 is, again, uncertain.  The market will continue to be responsive to the resolution of the tariff impasse, expectation about corporate earnings, the final resolution of Brexit between the UK and EU, the overriding talk in the trade press about (the need) for a market adjustment, the response by the FED and any additional stimulus provided by fiscal policy. (END)


Investors always seek signals indicating the future level of the S & P index. Some of the traditional indicators include the price / earnings ratio, future earnings, international trade and many others.  The following chart offers a graphic picture of a signal.


The variations in the above chart became more pronounced in early 2020.  It is difficult to interpret.  The following chart shows the period beginning 3 January 2020.


The signal offers a warning that the market is beginning a decline.  The week ending 31 January the signal indicated a potential decline as revealed in the signal = 1 on 2 February.  The two following weeks the signal returned to normal.  The signal on 2 March followed by another on 9 March exhibited the largest downward thrust in over two years.

The signal remained equal to ‘1’ until 6 April followed by another ‘1’ on 13 April. On 27 April the signal returned to ‘0’.

As long as the actual weekly close is below the lower forecast for two consecutive weeks expect the market’s decline to continue.  When the actual falls between the upper and lower bounds, i.e., the signal = 0, the market is not changing direction.

Using any Friday’s closing of the S & P index, a statistical forecast is calculated for the close on the next Friday.  A 95% confidence interval is calculated around the next Friday’s forecast. The interval is shown in the above chart as ‘Lower’ and ‘Upper.’

When the next Friday’s actual close occurs, it appears on the chart as ‘Actual.’  So the chart shows the forecast interval vs. the actual.

The signal is the actual close compared to the forecast interval (Upper and Lower) and is shown in the chart as the yellow line ‘Signal.’  Note on 9 February 2018 the actual close  was below the ‘Lower’ forecast signalling a change in direction of the S & P index. The up and down movement in the ‘Signal’ continued until 13 April when the actual Friday close returned to close between the ‘Lower’ and ‘Upper.’  The upward trend in the weekly closing price of the index resumed until 19 October.

On 19 October 2018 the actual close fell below the forecast ‘Lower.’  The upward trend in the S & P index stopped to be replaced by a declining trend until 2 January 2019. Beginning on 2 January 2019, the upward trend resumes and continues.

How do investors use the signal?  When the ‘Signal’ equals one, then a change in direction of the index follows.  Do not buy the S & P index until the ‘Signal’ returns to zero.  The signal returned to zero on 20 April.

Investors can buy the S & P index anytime the signal is zero.

In summary, the ‘Signal’ indicates a period of a ‘market correction’ when there is a pause in the upward and now downward momentum.

The signal is below the lower forecast limit for two consecutive weeks. This indicates the upward trend in the market has ceased.


Another slant on the market’s direction uses a measure of uncertainty.


The chart above shows the same time period as the preceding chart showing the ‘Signal.’

The uncertainty index is the difference between the ‘Upper’ and ‘Lower.’  From 5 January 2018 uncertainty slowly drifted up settling around 125.  Then, uncertainty abruptly increased on 11 January 2019. It fluctuated around 275 and slowly drifted down.  Uncertainty accelerated rapidly beginning in early April.  The whipsaw trace since early April suggests the market responds the any tidbit of both positive and negative information. Currently, the market seems to only respond to announcements about progress in the development of virus vaccine.

Note the three periods in the uncertainty index. The first is a slow but stable slight increase from January 2018 to January 2019.  In January 2019 the index shot up from near 130 to 290 followed by a slow decline until January 2020. The sharp decline to 145 in January 2020 followed by a flat period around 145.  Then the virus effect became obvious in late March.

The index increased significantly with large week to week variations. The index seems stabilize at 260 beginning on 15 May.

The exact cause of the increase in uncertainty is unknown. The increase is most likely the result of a number of factors – Trump, China, Brexit, …. clearly the virus and the lack of the federal government’s response, and now the prospects for the development of a vaccine.


The following chart shows the weekly closing price of the S & P 500 for 2018


The following table shows the summary of the actual and forecasts and actual and forecast for the weekly change in the S & P 500 for the year 2018.


The total of the actual forecasts of -144.56 is not so good. One of the reasons for the negative value is the forecasting algorithm is that it is trend following. One of its characteristics is that the statistical methodology identifies statistically significant deviations from the trend after they occur.  Unknown and unforeseen events will occur in the future, which will affect the underlying trend; but, their effects can only be identified after they occur.

Such unknown effects occurred during 2018.  Most of the effects originate in the White House, i.e., trade wars, brow-beating and name-calling, etc. and the market reacts.


The S&P 500 began the year on 30 December 2016 at 2338.83.  On 29 December 2017 the S&P 500 closed at 2673.61.  The total change in points = 334.71, which is a 14.31% increase.

Forecasting of the weekly closing price of the S & P 500 began for the week ending 23 February 2017.  The chart below shows the weekly actual closing price.


The following table shows the summary of the actual and forecasts and actual and forecast for the weekly change in the S & P 500 for the year 2017.


Last updated on 6 April 2020 using data for the week ending 3 April.

Source: FRED, S&P 500, weekly, aggregation method – end of period, and the author’s calculations.


(NEW) What is clear is that the trade war continues unabated, the FED is doing its best to read the tea-leaves about the effect on GDP from all the variables affecting the U.S. economy, and Trump and the White House refuse to accept any responsibility for the turmoil.

What is clear is that the future path of GDP is uncertain.  The daily and weekly swings in the S & P index reflects the uncertainty.(END)


The future GDP growth rates shown below reflect the underlying/historic pattern of GDP.  The effects of the massive tax cut directed towards high income individuals and corporations has dissipated.


(NEW) The graph above reflects the revised GDP released on 19 January.  The revisions include a reduction in the 2Q growth rate from 3.1% to 2%.  The forecasts show a constant growth rate of 2.4% for 2020.  At this time there is no indication that the growth rate will become negative, but the forecasts cannot be good news for individuals and firms making investment decisions.

The Bureau of Economic Analysis releases its first estimate at the end of the month following the end of the quarter (in April) and its second release during the last week at the end of the second month following the end of the quarter (in May).

With all the troubles Trump has instigated, the forecast for GDP growth rates as the November 2020 election approaches cannot be good news.

The trade press and every forecaster offers their version of what the future holds.  None reflect any pattern in the monumental effects on all the macroeconomic series available.  Hence, all the forecasts are the forecaster’s best guess. The forecast for the 2Q2020 shown below is a guess.  Future forecast of the quarterly GDP will be offered after the historical patterns in the relevant series are established.

Analysis updated on 15 May reflecting data thru 1Q2020.

SOURCE: FRED, series – GDPC1


The graph below shows the actual and forecast of the 12 month percent change in CPI.    The actual decline will have some practical lower limit absent a recession.  The FED’s lower target rate is 2%.  The continued decline in prices results in consternation at the FED.

The statistical model includes 25 intervention variables, including two seasonal pulses.  For those who are interested the R² = .9999.

A word of caution:  Trump’s continued harangue about the current trade deficit with China and his imposition of additional tariffs (in the last few days, the White House indicates the latest increase will be moderated) will eventually produce an increase in the CPI.  The impact on the CPI will depend on a host of factors where the magnitude is unknown at this time.  In the end, American consumers will shoulder the price increases.

The forecasts suggest that an increase in inflation is not a threat.   The FED’s actions and printed commentary indicate it is either overly concerned about immediate inflationary pressures or desires to slow the overall economy in an attempt to forestall any future inflation.

Given the uncertainty surrounding the national and state economies, the forecast is only for one period ahead – May 2020.  The forecast is for the CPI to increase by .43% next month following a .38% increase from the previous 12 month value.


The Bureau of Labor Statistics releases the previous months estimate of CPI in the middle of the current month.

Updated on 15 May 2020 using data thru April 2020.



The U.S. unemployment rate continues to flatten at near 3.5%.  The graphic pattern suggest that normal job turnover, along with new hires and layoffs, result in the 3.5% minimum.

The FED has to contend with the small monthly fluctuations in the unemployment rate, expected small increases in inflation and slow increase in wages when it considers another rate increases.  Furthermore, there is continued chatter from the political class (lead by Trump last year )that interest rates should increase to increase the earnings of savers. This argument is most mysterious, since the traditional Keynesian view is that investment decisions are influenced by the spread between the expected rate of return and the interest (cost of funds) rate.


The flat line forecasts of the unemployment rate at 3.5%  suggests the FED will view the outlook as non-treating to the U.S. economy.

The FED is under increasing pressure from Trump. The outlook for the U.S. economy is becoming more questionable every day.  The spread between the U.S. Treasury 10 year constant maturity and the 3-MO bill rate is negative.  This is not a good sign.  The longer the spread is negative the greater the chances that the economy enters a slowdown.


The flat forecast of the unemployment rate reflects the estimated statistical equation that summarizes the historical data. The equation is very simple – stationary using first differences and only one pulse intervention variable. It does not include any AR or MA terms.  The future forecast values are the actual unemployment rate for the previous month.

The effect of the virus is not detectable on the unemployment rate – YET.  However, reported virus infections continue on the upswing, and it is only a question of time before the magnitued and duration of its impact on all the world’s economies can be measured.

Updated on 14 March 2020 using data thru February 2019.

SOURCE: The unemployment rate is calculated to six decimal points using data from FRED, series – CLV16oV and LNS13000000.


The average monthly change in U.S. payroll employment for the 12 months ending December = 176,000.

The month-to-month change in total payroll employment is more of interest and more closely watched that the monthly total.  The following graph shows the actual monthly change from January 2018 thru  December 2019 and the monthly forecasts January – June 2020.

The statistical estimate for 12 month change in January 2020 is 202,000.

The FED watches the monthly change in payroll employment closely.  It, along with the unemployment rate and the forecasts of GDP,  guide the FED in its interest rate decisions.  The following graph shows normal month-to-month changes in payroll employment along with the expected monthly changes during the remaining months thru August 2020.


The FED reduced interest rates on an emergency basis last week in response to the carnoravirus.  Monetary policy, i.e. a reduction in interest rates, will have little effect in the short run. The reduction is insurance.

Updated on 14 March 2020 using monthly data thru February 2020.

SOURCE: FRED, series PAYEMS, seasonally adjusted