In the original comment of April 4, 2009, it has been forewarned that the annual revision of inflation adjusted Personal Consumption Expenditure (real PCE) around the end of July of 2009 may change the pattern of real PCE compared to the old pattern based on the data before the annual revision. Now the annual revision is made public. Indeed the pattern of real PCE has changed to the degree of requiring our attention. In the first graph posted at right monthly values of revised real PCE are plotted as the green curve in log-10 scale with the scale shown in the left margin of the graph as green letters. Similarly the old real PCD before the annual revision is plotted as the red curve with its log-10 scale plotted at the right margin of the graph in red letters. It should be noted that the revised PCE data are calculated with the price index of year 2005 set as 100, whereas the real PCE data before the annual revision was calculated using the price standard that set the price index of year 2000 as 100. Thus the comparison of the absolute values of the green curve with that of the red curve becomes meaningless, but only the comparison of relative patterns of those two curves counts.
The red old real PCE curve showed that it had bottomed out in the fourth quarter of 2008, and was bouncing back in the first quarter of 2009 and beyond. However, the green revise real PCE curve shows that it is still declining albeit at a reduced pace. The sharp downward spike of December of 2008 was rather an aberration. After forming a slightly upward shoulder in the early months of 2009, the new real PCE curve is in a downward channel again though the slope of the decline has slackened a lot compared to the earlier decline from the spring to the end of 2008. This outcome of revised real PCE is consistent with the view of the Federal Reserve Board (FED) and many other economists that the economy has slowed its pace of decline, but is still declining through the first half of 2009. Some Wall Street experts, probably misled by the pattern of the old real PCE data, had claimed that the recession had already ended in the first quarter of 2009 in order to give credibility to the stock market rally started in March of 2009. Those Wall Street experts are not aware of the fact as pointed out in Comment 64 that after the nadir of a recession a strong stock market rally, either as the genuine first leg of a new bull market or just as a powerful bear market rally, had taken place in all the past recessions since 1955. The start of those stock market rallies had never waited for the official end of the recession.
To estimate when the nadir of this recession will be reached and when this recession will end, we need to look beyond real PCE and jump into the analysis of real GDP. In the second graph that is posted at left, annualized values of real GDP are plotted in quarterly interval as the black curve. The red curve is the year-to-year % comparison of real GDP with the position of each horizontal bar shifted toward left by two quarters. When the red curve is below the dark green +1.0% line, the economy is in a recessionary environment. The nadir of the recession is the lowest point of the red curve. Those are the definitions about recessions adopted in article 12. Monthly values of nonfarm payrolls are plotted as the green curve in the upper portion of the graph for later use.
The last horizontal bar on the red curve is at the fourth quarter of 2008 and is labeled as R in the second graph. R is the % comparison between the real GDP of the second quarter of 2009 (labeled as E) and the real GDP of the second quarter of 2008 (labeled as A). The position of R is determined by shifting two quarters toward left from the second quarter of 2009, that is, at the fourth quarter of 2008. To compute the red bar at the first quarter of 2009, we need to use the real GDP of the third quarter of 2009 (not yet available) and the real GDP of the third quarter of 2008 (labeled as B). From A to B real GDP had declined by -2.7% in the annualized rate. If the performance of the real GDP of the third quarter of 2009 is better than -2.7% (annualized), the red bar at the first quarter of 2009 will be above R. Assuming no further disastrous showing of real GDP after the third quarter of 2009, R at the fourth quarter of 2009 will be indeed the nadir of this recession. As mentioned already, in Comment 64 it has been pointed out that a strong stock market rally had always taken place right after the nadir of a recession in the past eight recessions since 1955. Such rallies typically last about six months and then had come to a turning point. The further performance of the stock market beyond the turning point depended on the exact ending of a recession and how strong the recovery after the recession had been as discussed in Comment 70. The current strong stock market rally that has started in March of this year seems to conform to the observation of Comment 64 if R is the nadir of this recession.
According to the definition adopted in article 12, this recession will end when the red curve in the second graph moves above the +1.0% line. Since the last bar R on the red curve is located at the fourth quarter of 2008, in principle this recession can have ended in the first quarter of 2009 if the red bar at the first quarter of 2009 will move above the +1.0% line. The red bar at the first quarter of 2009 will be calculated from the real GDP of the third quarter of 2009 and the real GDP of the third quarter of 2008 labeled as B. Thus from B and E we can calculate the performance of the real GDP of the third quarter of 2009 that is required for this recession to end in the first quarter of 2009. It turns out that the real GDP of the third quarter of 2009 needs to rise +17.5% (annualized) or better in order for this recession to end in the first quarter of 2009. Such a crazy growth of real GDP in the third quarter of 2009, of course, is out of question, so the claim of some Wall Street experts that this recession has already ended in the first quarter of 2009 can be dismissed easily. How about the case that this recession has ended in the second quarter of 2009? Calculating from black bars labeled as C and E, the average annualized growth rate of the real GDP in the second half of 2009 needs to be +5.9% or better to bring this recession to an end in the second quarter of 2009. The most optimistic estimate about the average growth rate of real GDP in the second half of 2009 that we have heard from Wall Street is in the range of +3.0% to +3.5%. We deem the +5.9% figure, though not impossible, highly unlikely so the chance of this recession ending in the second quarter of 2009 can be ignored, too. When it comes to the third quarter of 2009 as the ending quarter of this recession, the barrier is lowered substantially. From the calculation using D and E, the average growth rate of real GDP in the coming three quarters needs to be +1.7% or better for this recession to end in the third quarter of 2009. Using E alone, we can calculate that the average growth rate of real GDP in the coming four quarters needs to be only +1.0% or better for this recession to end in the fourth quarter of 2009. It is very likely that this recession will end in the third or in the fourth quarter of 2009.
The green nonfarm payroll curve in the second graph is in general a lagging indicator. However, in this case it can serve as a useful indicator about when this recession will end before the red bars needed can be calculated. In the past recessions since 1955, the nonfarm payroll numbers had already turned up or turned to a sideway trend at the quarter after the one that a recession had ended. Therefore, if the green nonfarm payroll curve does not level off by November of 2009, the chance for this recession to end in the third quarter of 2009 will diminish substantially. If the green nonfarm payroll curve is still falling gently near the end of the first quarter of 2010, then the chance for this recession to end in the fourth quarter of 2009 will be threatened, too. Even if this recession ends in the third quarter of 2009 and the green nonfarm payroll curve almost levels off in the fourth quarter of 2009, this does not guarantee that this recession is a V shaped one with vigorous recovery. To assure a V shaped recession the average growth rate of real GDP in the second half of 2009 needs to be in the range of +3.0% to +3.5%, and even better in 2010. However, such robust recovery carries substantial risks as will be discussed below.
In Comment 31 it has been pointed out that prior to the era of runaway trade deficits the rule of thumb for a stable monetary policy is to keep the short-term interest rate at par with the average growth rate of nominal GDP; nominal means without the inflation adjustment. When the short-term interest rate fell below the average growth rate of nominal GDP, higher inflation rate followed, requiring the short-term interest rate be lifted above the average growth rate of nominal GDP to curb the inflation. If the average growth rate of real GDP to be +3.0% to +3.5% or better after the end of this recession, the average growth rate of nominal GDP will be about +6.0%. Thus the short-term interest rate needs to be raised to around +5.0% at least by the end of 2009 to avoid the emergence of a heightened inflation rate if trade deficit is kept at the current level. It will be very difficult for FED to do this kind of rapid turn around from the current near-zero short-term interest rate policy. Then there is the problem of the huge government budget deficit that is equivalent to a giant stimulus package for the economy. In order to avoid the emergence of a hyper inflation under the scenario of +3.0% to +3.5% average growth of real economy and without the help form another phase of runaway trade deficit, the budget deficit needs to be brought down below 500 billion dollars in the fiscal year of 2010. Thus a hyper inflation is inevitable if the real economy is to grow like +3.0% to +3.5% in the second half of 2009 and beyond without the help from the runaway trade deficit. In that case the economy will fall into another deep recession to make the dismal prediction of a W shaped recession come true.
If the fourth phase of runaway trade deficit is allowed to take shape, the emergence of a hyper inflation can be suppressed by letting the onslaught of inexpensive foreign made goods do the trick even under the condition of 3.0% to 3.5% average growth rate of real GDP in the second half of 2009 and beyond. However, the size of the runaway trade deficit needs to rise to the level to make the size of the third phase of runaway trade deficit of Bush era pale by comparison. This is due to the inability of both FED and U. S. Government to withdraw massive liquidity and stimulus already injected into the economy in time. The giant size of the runaway trade deficit in the fourth phase will inevitably generate a monstrous bubble the burst of which will literally push the global economy into a depression.
The grand liquidity squeeze, which has been induced by the wane of US trade deficit, has spurred the first financial firestorm in August of 2007, followed by a series of such financial firestorms that eventually has led to the meltdown of Wall Street in the fall of 2008. This series of incidents are followed on this website through a series of comments most of which carry a subtitle "Tracing the liquidity squeeze", starting from Comment 46. In response to the evolving financial and economic crises FED, US Treasury Department, FDIC, and Federal Home Loan Banks have installed ever more grandiose and aggressive bailouts, stimulus packages, loans, guarantees, outright purchases of toxic assets, prime mortgages and even long-term US Treasuries. Through the bailout of losers like AIG in the gambling field of OTC derivatives, US Government is even handing out the full amount of winnings of the gambling game to the counter parties of AIG. Total amount of money handed out by US Government, including future commitments and guarantees is said to have reached close to 13 trillion dollars already. This amount should be compared to the total debt of non financial sector about 30 trillion dollars before August of 2007. It is natural for many, including Chinese Government, to voice concern about the possibility of a future hyper inflation when the recovery takes place and if the mind boggling amount of liquidity injected into the market by US Government fails to be withdraw with a lightening speed. Thus we start our examination from the inflation front, trying to catch any early signs of trouble.
Monthly values of Consumer Price Index including all items, called "CPI-all", are plotted as the green curve in the graph at the right. Monthly values of Consumer Price Index less food and energy, called "core CPI", are plotted as the red curve. The green CPI-all curve fell sharply from September to December of 2008, but has turned up in January and February of 2009. The sharp fall of CPI-all was due to the collapse of crude oil price as the component of energy goods and services within CPI-all indicated. However, the rise of CPI-all in early 2009 cannot not be attributed to the rise of crude oil price since crude oil did not rally in a significant way in January and February, but has started to rally strongly only in March. The red core CPI curve flattened out from August to December of 2008, and then has risen in January and February of 2009. The rise of CPI-all in early 2009 is thus the result of the strengthening core CPI. The rise of core CPI stirs the anxiety whether the nadir of the recession has already passed and our attention now should be switched to inflation watch. In order to investigate this possibility we turn our attention to the largest component of GDP, personal consumption expenditure next.
In the second graph at the right, the top group of bluish lines are monthly values of inflation adjusted Personal Consumption Expenditure, called "real PCE", expressed in the scale of log-10(billion dollars) that are shown at the left margin of the graph. The lower group of reddish lines are monthly values of real PCE less food and energy, and their logarithmic scales are shown in the right margin of the graph. The most up-to-date data of real PCE can be read from the graph as follows: Start from the lower end of the dark blue line, follow it up and branch into the light blue line, and then branch into the green line. Similarly the way to read the most up-to-date data of real PCE less food and energy is to start from the lower left end of the red line, follow it up and branch into the brown line, and then branch into the orange line. Both real PCE and real PCE less food and energy have been falling sharply starting from early summer of 2008, but have leveled off in the 4-th quarter of 2008. They have staged some bounce-back in January and February of 2009. The average of real PCE in the 4-th quarter of 2008 was substantially lower than that of the 3rd quarter, contributing substantially to the miserable showing of real GDP of the 4-th quarter of 2008 when compared to the result of the 3rd quarter of 2008. According to the trend of real PCE as shown in the graph, the average of real PCE of the 1st quarter of 2009 will be at par or slightly better than that of the 4-th quarter of 2008, indicating that when real GDP data for the 1st quarter of 2009 comes out at the end of this April, quarter-to-quarter % change will be substantially better than the miserable result of the 4-th quarter of 2008. This observation raises the hope that the nadir of the current recession may have been reached already in the 4-th quarter of 2008.
Before we can proceed further in analyzing the progress of the recession, we need to make clear what the nadir of the recession means. If quart-to-quarter % change of real GDP is a relatively smooth curve and is plotted through a recession, the lowest point of that curve is defined as the nadir of the recession. Unfortunately as demonstrated in article 12, quarter-to-quarter % change curve of real GDP is very jumpy and some kind of smoothing method needs to be applied in order to derive some useful information from such a wildly gyrating set of data. In article 12 year-to-year % change of real GDP is used as the measure of smoothing. Suppose this year-to-year % change curve reaches its lowest point in a recession at quarter P. By pushing P back by 2 quarters, that is, quarter (P - 2) is defined as the nadir of the recession. The process of "push back by 2 quarters" is necessary since any smooth-out process is done at the price of timeliness so that year-to-year comparison is looking at the average condition of 2 quarters ago. The beginning and the ending of a recession can be defined similarly using the year-to-year % change curve. Suppose the curve drops below +1.0% at quarter B, push it back by 2 quarters, and the quarter (B - 2) is defined as the beginning of the recession. If the curve jumps above +1.0% line at quarter E, the recession is then defined as ended in the quarter (E - 2). The results of this elaborate definition of recessions adopted in article 12 coincides well with the declarations about the beginnings and endings of recessions from National Bureau of Economic Research, the official arbiter of economic cycles in US. Some may ask why the deliberate definitions of recessions are needed when the official declarations of National Bureau of Economic Research are available. The declarations of the bureau is subjective in nature. The declarations are based on the majority vote of a board that consists of eleven prominent economists. Each economist on the board is allowed to use any of the criteria he/she sees fit, and those criteria are usually not revealed to the public. Thus it is difficult to analyze the progress of any recession based on the vastly delayed declarations from the bureau in a timely manner. It should also be noted that there is a widely used folklore type definition of a recession as two consecutive negative growth quarters of real GDP. The beginnings and the endings of recessions from this folklore type definition diverge widely from the declarations of the bureau. Furthermore, the folklore type definition misses some important recessions. For example, the 2001 recession following the collapse of the Clinton era giant stock market bubble will be deemed not a recession at all since there were no two consecutive negative growth quarters in real GDP throughout that time span.
With the notion of the nadirs, the beginnings and the endings of recessions clarified, we can now proceed to guage the progress of this recession. Let us start from the assumption that real PCE will move in the sideway trend far into the future, and as the result quarter-to-quarter % change of real GDP, starting from the 1st quarter of 2009, will be near 0% for a long time to come. A straightforward calculation shows that the year-to-year comparison curve will reach its lowest point, -1.75%, in the second quarter of 2009, indicating that the nadir of this recession has been reached in the 4-th quarter of 2008. The year-to-year comparison curve will rise to 0% by the 4-th quarter of 2009, and will stay at 0% forever. This character will deem this recession as an open ended "L" shaped recession that we call "Japan Syndrome" just as Japan has experienced since 1995. In this case inflation will not be a problem at all. Let us make the condition somewhat more optimistic and assume that both real PCE and real GDP will rise gradually from here. The nadir of this recession is still at the 4-th quarter of 2008 under this condition. Suppose the year-to-year comparison curve rises above the +1.0% line at the 2nd quarter of 2011, indicating that this recession will end at the 4-th quarter of 2010. According to article 12, this recession has started in the 1st quarter of 2008, whereas National Bureau of Economic Research has declared that this recession has started in December of 2007. Thus this recession under this condition will last for 12 quarters, and will be classified as an extended "U" shaped recession. This means that in the case of the extended "U" shaped recession scenario, FED must withdraw massive amount of liquidity injected into the market previously with a lightening speed starting from the beginning of 2011, and US Government must cut back on its massive budget deficit with equal vigor even at the expense of many social agenda installed by this administration, otherwise the hyper inflation dragon will lift its ugly head. What happens if things are a little more rosy and let us assume that the year-to-year comparison curve passes the +1.0% line in the second quarter of 2010, meaning that this recession will end at the 4-th quarter of 2009. In this case this 8 quarter long recession will be classified as a "U" shaped recession, and the necessity of the massive reversal of policies by FED and US Government will be pushed up to early 2010, just less than a year away. The most rosy scenario that we prepared to entertain is to assume that real PCE and real GDP will zoom upward form here. In this case the nadir of this recession is still at the 4-th quarter of 2008. Suppose the curve of year-to-year comparison of real GDP crosses the +1.0% line at the 3rd or the 4-th quarter of 2009. This means that this recession has already ended in March this year, or will end in this April or in coming May. Under this scenario this 4 quarter long recession will be deemed to be a deep "V" shaped recession. If that is the case the extremely expansive FED and US Government policies at present are not only unnecessary, but are outright dangerous, and the emergence of a hyper inflation will be inevitable. In the aftermath of fighting the hyper inflation, US economy will fall into another deep recession soon.
The scenarios discussed so far are all based on the assumption that the nadir of this recession has already been reached in the 4-th quarter of 2008, and the worst case is that real PCE and real GDP continue their sideway movement from here. Those assumptions, however, is not carved into stone, because there is an important piece of data showing discord with this rosy view. The data is about the employment situation. The monthly numbers of nonfarm payrolls are plotted as the blue curve in the first graph posted in this comment, with its linear scale shown at the right margin of the graph. The blue nonfarm pay roll curve peaked in December of 2007, showing a perfect timing with the entrance into this recession at the 1st quarter of 2008 whereas the peak of real PCE in the second graph lagged behind by 5 months to the onset of the recession, though in earlier recessions nonfarm payroll numbers usually lagged behind real economic events compared to the timing of real PCE. After its peak at December of 2007, the blue nonfarm payroll curve accelerated its sharp decline until the most recent data of March, 2009 at the time when real PCE is leveling off as discussed before. The sharp conflict between the nonfarm payroll number and real PCE is now apparent to any objective observers unless nonfarm payroll number shows sharp break and level off in the next month or so. If not, there are two possible ways that this conflict may be resolved. The first possibility is that real PCE will resume its sharp fall soon and conform to the sharp drop of nonfarm payroll numbers. Another possibility is self evident from the curves shown in the second graph. Bureau of Economic Analysis is the government agency that compiles PCE and GDP data. The bureau makes annual revision of those data around the end of every July, in coincidence with the release of June PCE and the second quarter GDP data. Before the July 28, 2007 annual revision, real PCE curve in the second graph was thought going along the dark blue line until the May, 2007 data point. A straight line labeled as "D" can be drawn on this logarithmic graph, indicating that real PCE was growing with a constant rate along the line "D" since 2003 without showing any sign of slow down. However, the July 28, 2007 annual revision had revised down real PCE to the light blue line shown in the graph. The trend line "D" was violated already in the second half of 2005, indicating that the growth rate had slumped by a substantial amount. The August 4, 2008 annual revision put the real PCE curve down further to the green curve in the graph, implying that the growth rate of real PCE had suffered another sharp slow down since early 2007. The same massive revision may be repeated at the time of next annual revision around the end of this July, and the leveling off of real PCE discussed earlier may just become a mirage. In those worst case scenarios, the nadir of this recession, of course, is far from being reached. Considering panicking reactions of FED and US Government in recent days, it seems to us that they are bracing for the worst case scenarios and are hoping to avoid the catastrophe by blanketing the economy with unlimited amount of money created from nowhere.
US stock market has staged a powerful rally from March 6 low. Many are wondering whether the bear market bottom is already formed and we are witnessing the first leg of a long lasting bull market. To answer this question we must first distinguish bear market bottoms from bull market tops. As discussed before bull market tops tend to precede the entry point of a recession by a few months in average, and thus promote the claim that stock market is a leading indicator. However, bear market bottoms always come after the nadir of recessions but not before as shown in Comment 64. All the recessions studied in Comment 64 except the 2001 one, are "V" shaped recessions. For those "V" shaped recessions the average lag time of bear market bottoms to the nadirs of recessions is one quarter. The 2001 recession stands out as a "U" shaped recession. The nadir of the 2001 recession was reached in the second quarter of 2001 and the recession ended in the 1st quarter of 2002 to the 2nd quarter of 2002, making the recession 5 quarters long. US stock market did rally up as the recession came to the end, but fell shortly after as the recovery was proven to be wobbly, a trade mark of a "U" shaped recession. US stock market eventually reached the bear market bottom around the end of 2002 to the early part of 2003, and Dow Jones Industrial Average sank to a much lower level than the low formed in September of 2001. Borrowing from the experience covered in Comment 64, we may say that the current rally can be the first leg of a bull market if only this recession is a deep "V" shaped one. Unfortunately as discussed before, if this recession is indeed a "V" shaped one and the economy is already recovering strongly, then the extremely expansive policies of FED and US Government are not only wrong but poisonous, and a hyper inflation will follow. As the hyper inflation takes hold, a more devastating recession and another nasty bear market will follow soon. In any event the best case is that the current rally is the first leg of a short bull run, but not the first leg of a long lasting bull market. If the current recession is a "U" shaped one instead, the bear market bottom apparently has not been reached and the current rally is just another powerful bear market rally. If this recession is an extended "U" shaped one, the bear market bottom will be pushed out farther into the future. If this one turns out to be an open ended "L" shaped recession and the "Japan Syndrome" takes hold, then by definition the bear market bottom cannot be seen as far as eyes can reach. In the worst case scenario as discussed in the previous paragraph, we just do not know where the bear market bottom is and when it will be reached. In that occasion it is probably worth to look into the case of The Great Depression. From the 1929 top to the 1932 bottom, Dow Jones Industrial Average had lost almost 90 % of its value. To apply the same degree of devastation to this recession, Dow Jones Industrial Average needs to drop to around 1430, and will make the bottom line discussed in Comment 61 looks like a paradise.