Comment 24: Is a recession looming? 2. Why oil price has lost its punch. How about the hurricane? (Sept. 8, 2005)

As the price of crude oil marches continuously higher with the price of gasoline advancing in lock step with the price of crude oil, many financial experts have been predicting an imminent recession due to the energy crisis for quite a while. First they said that a recession will hit when the price of crude oil reaches $40/barrel, then $50/barrel, $60/barrel, $80/barrel and so on. The last price $80/barrel is derived from the peak oil price during the energy crises of 1970's, adjusted for inflation from that time to the present. Astute readers have probably noticed already that in the last energy crises economic recessions had already hit before the oil price reached its peak equivalent to $80/barrel today. Those readers must be asking why the oil price shock has lost its punch today to induce an economic recession? The first part of this comment will be studying this important question. As hurricane Katrina has devastated a large portion of the coasts of The Gulf of Mexico and caused substantial human sufferings along with heavy economic losses, many analysts have again jumped to the proclamation that an economic recession will now arrive due to this natural disaster. In the second part of this comment we will look into this subject.

Energy consumption is only a small fraction of GDP (Gross Domestic Production) used widely to gauge the performance of an economic entity. In order for the energy price to influence the flow of the whole economy, the price movement of energy must be reflected on the price movements of other components of GDP. Various price indices are plotted in logarithmic scales in the figure. The absolute values of those data points are not relevant; those curves are shifted vertically to make them fit in one graph. The shape of each curve and the correlations among the shapes of the curves are the focal points here. The black dots are the price index of energy component of the non-durable goods that in turn is a component of the personal consumption. The first energy crisis in 1973 to 1974 is clearly visible as a sharp rise in the curve. The second energy crisis occurred during 1979 as shown as another sharp rise in the curve. The most recent up leg in the oil price has begun since 2002; the energy price index has hit a temporary bottom at the first quarter of 2002. Since then the energy price index has risen over 70% in three years with a roughly constant rate of increase, that is, forming an approximately straight line in the logarithmic scale. The green dots are the price index of residential fixed investment, that is, the new homes and apartments built during the accounting period. The price index of the residential fixed investment has synchronized with the energy price up to early 1980's. From there the energy price index stagnated for nearly 20 years and has only turned up strongly from 2002 as mentioned above. However, the price index of the residential fixed investment grows with a constant rate of like 5% a year since early 1980's, implying the decoupling from the energy price index. The blue dots represent the price index of service. It displays a similar behavior as the price index of the residential fixed investment. The decoupling of the price index of service from the price index of energy since early 1980's is also evident. Since service has become the largest component of overall personal consumption, the price index of overall personal consumption, the red dots, displays a pattern similar to the price index of service. Durable goods consumption composes a rather small fraction of the overall personal consumption and its price index is not shown in the graph; the price index of durable goods has peaked in 1996 and has been falling steadily from that time on. The reason of this behavior of the price index of durable goods is quite obvious. It is due to the onset of information technology and internet age. The prices of information technology gadgets, including computers, are falling rapidly due to the continuous advance in computer related high technologies. This behavior of durable goods price has offset the steadily rising service price and has moderated the rise of the overall price index of personal consumption. In recent 20 years, the price index of the overall personal consumption has risen only 58%, whereas the price index of service has risen almost 100%. The non-residential fixed investment, the price index of that is represented as purple dots, is an important component of GDP since it usually gyrates strongly compared to the more dominant but more resilient personal consumption. In other words the ebbs and flows of non-residential fixed investment often dictate the performance of the whole economy. The price index of non-residential fixed investment synchronizes strongly with the price index of energy up to 1995. Then it has parted from the price index of energy. This parting is even more striking compared to other price indices discussed so far since it started to fall from its 1995 peak. The major part of the non-residential fixed investment is the investment in equipments and software. The price index of this component behaves similar to the price index of durable goods discussed before; it is synchronized with the price movement of energy until 1995, then its price has been falling steadily since that time. For the price index of the smaller component of non-residential fixed investment, the investment in factories follows the movement of the price index of the residential fixed investment as expected.

The analyses of the previous paragraph indicate the decoupling of general price indices from the price movement of energy. The decoupling takes two very different forms. The price index of service that dominates the component of personal consumption and the price index of residential fixed investment have risen with a steady rate since the onset of the globalization scheme in early 1980's, whereas the price of energy stagnated for 20 years and only turned up strongly since 2002. The price index of non-residential fixed investment has synchronized with that of energy until 1995, and then has started to fall in spite of the strong upturn of the energy price since 2002. With this kind of decoupling from the price of energy, we can confidently say that the energy price has lost its punch in influencing the ups and downs of the whole economy. The fear of a recession due to the rising energy price is apparently not warranted. Looking back to the years preceding the first energy crisis in 1973 - 1974, we can see that it was the general price index leading the price of energy; only with the Arab oil embargo energy price overshot the general price index in the period from 1973 to 1974 and created the first energy crisis. Even without the oil embargo, the rising price inflation in the general sector would have had induced a very high inflation expectation and a recession. Before the second energy crisis of 1979, the general price and that of energy price were rising in tandem. The second energy crisis, triggered by the fall of Shah of Iran, though pushed up the price of energy sharply, had only a limited impact on the general price index as shown in the red curve of personal consumption price index in the graph. The recessions around 1980 and 1982 were the result of failed economic policy of Carter administration and then the iron fist monetary policy of FED to uproot the deeply entrenched inflation, though conveniently blamed on energy crisis by politicians and economic spin-doctors. In summary it is very dangerous to anticipate a recession based on the movement of energy price in the current globalization scheme under that the decoupling of general price movement from that of energy is proceeding strongly..

Hurricane Katrina has caused numerous damages and exhorted heavy human tolls along with massive disruption in oil and gas production. It also has created transportation nightmares. Many analysts have immediately jumped on the bandwagon and proclaimed that the hurricane will finally push the economy into a recession that the persistent FED tightening and the rising oil price have failed to deliver so far. However, we need to be less emotional and analyze the economic facts objectively to see what impact hurricane Katrina may have on the whole economy. To tackle this problem we need to start from the definition of an economic recession. It is reasonable to define a recession as a prolonged period of diminished growth rate or sometimes a negative growth rate in Real Gross Domestic Production (Real GDP). Economic recessions defined that way can easily be spotted by plotting Real GDP in a logarithmic scale as done in the figures quoted in Comment 23. We do not need the artificial definition that classifies a recession as two consecutive quarters of shrinking Real GDP. According to that artificial definition of a recession the economic slow down from 2000 to 2003 after the burst of the stock market bubble was not a recession since in the revised Real GDP data there are no two consecutive quarters of declining Real GDP during the period from 2000 to 2003, but in a logarithmic scale plots of Real GDP as in the second graph of Comment 23, that stretch stands out clearly as a recession. With the definition of a recession settled we need to look into the ways GDP is compiled to understand how to assess the impact of the hurricane on GDP.

As the name Gross Domestic Production suggests GDP is to count the total domestic production within a fixed time period, usually within a quarter and expressed as an annual amount by multiplying a factor of 4 on the quarterly number. Let us consider a product out of a factory that costs $1. It is no problem to count this $1 and include it in GDP. However, when this $1 object moves through the retail chain, the costs of handling accumulate and the value of the object increases. All those increased values are also the domestic products that should be included in GDP. It is quite tedious to count all those added values as the object moves through the retail chain. Suppose the object is finally sold to a consumer for $10. It is clear that it is much easier to count this $10 as the final value of this object and to include it in GDP than to count the value created in each stage when the object change hands. This is exactly what is done in GDP. The total sum of final sales to consumers counted that way is called "The Final Sales" or "The Personal Consumption", and accounts about 75% of the GDP of The United States of America. Though this way of counting the final sales to consumers is ingenious, it does contain a hole. Some of the objects sold domestically are produced in foreign factories, so the first $1 of the object, if it is imported, should not be counted as domestically produced. Instead of trying to decide what fraction of the final sales is from domestic origin, a very difficult task indeed, we just take the final sale as it is but subtract the total amount of imports, from the data of the custom, later on. This subtraction of imports in the GDP account has caused a lot of confusion even among economists and financial analysts, and leads them to think erroneously that imports contribute negatively to GDP. They think simplistically that by importing this $1 object $1 is subtracted from GDP, but they fail to recognize that this subtracted $1 is already included incorrectly in the final sales and the subtraction is simply correcting this error. If this $1 object is not imported, $1 certainly does not need to be subtracted from GDP in the import category, but the final sale account will lose whole $10. This implies that the import actually will contribute positively to GDP rather than the erroneous version propagated by many financial media and some economics textbooks that imports are negative to GDP. Then we must consider the exports. Suppose an object worth of $1 comes out of a domestic factory and is exported. This $1 object will not be sold to a domestic consumer and will not be included in the account of final sales, though this $1 is clearly produced domestically. That is why GDP include the total amount of exports, again from the data of custom, as a positive contribution. This addition of exports has again generated the erroneous version of many economists and economic textbooks to consider exports as positive to GDP. It should be noted that by exporting this $1 object, the chance of generating extra $9 for GDP if this object is sold domestically is lost. In an extreme case, if an economic entity exports everything it produces but does not import anything, does GDP of the entity reach maximum and the entity prosper mightily as those economists try to let the general public believe? If an economic entity does this kind of foolish things, all the people of the entity will perish due to the lack of things to consume and the entity will be wiped clean from the face of the earth! From this simple account of imports and exports readers should get a sense why the largest trade deficit country, America, is booming and the largest trade surplus country, Japan, is suffering economically. Though the correct accounting of imports and exports is extremely important for us to understand the booms and busts in the globalization era, a detailed discussion is not the purpose of this comment; the subject will be covered in a later writing but impatient readers are referred to Article 2 for somewhat more detailed discussion about this subject.

The next important component of GDP is called the private fixed investments. This component consists of three sub-components, change of inventory, residential fixed investment and non-residential fixed investment. New inventories generated in the accounting period are the products not yet sold to customers so they are not included in the final sales, but clearly they are values generated domestically and need to be included in GDP. Residential fixed investment is the value of new single-family and multi-family residential units built during the accounting period. Non-residential fixed investment is the values of new factories built and new equipments and software purchased by businesses. The final component of GDP is the expenditure of the federal and local governments. It is important to note that wages and salaries, savings and other categories, which are usually considered to be wealth, are not directly included in GDP. Those things are actually already been accounted for in some form or another in GDP, and explicit inclusion of them in GDP will result in double counting of the same items. The total amount of GDP compiled from those components is called the nominal GDP or the current dollar GDP. The real GDP is obtained by dividing the nominal GDP by a general price index called the implicit price deflator. Currently this price index is set to be 100% for the year 2000. The real GDP presently in usage is also called 2000-chained GDP.

When it comes to analyze the effect of hurricane Katrina, the first thing we should note is that the loss of human life and the vast human sufferings are not counted in GDP; GDP is simply not designed to count such things. The properties damaged and lost are also not counted in GDP since they are not new productions during the accounting period. Also the losses of businesses are not counted in GDP. Only when the factors mentioned above start to influence the final sales, their effects on GDP will be noticed. Let us assume that about 2 million people are stripped of any income for the foreseeable future by the hurricane. For those more affluent evacuees, holding up in hotels and motels in nearby localities, the out of pocket expense may not be less than before the hurricane; many of them probably are paying the away-from-home expenses by using their credit cards. However, GDP does not distinguish the sources of personal consumptions, whether they come from wages and salaries, from savings or from borrowings, one dollar spent is one dollar contribution to GDP. Thus for those more affluent evacuees their contribution to the final sales may not be dented very much. For those less fortunate evacuees under the care of the government, we must assume that they have lost all their contributions to the final sales. Let us assume further that in average every evacuee loses $15,000 a year in the contribution to the component of final sales. The total loss to the component of the final sales is thus 30 billion dollars a year. The lost production of oil and gas in the gulf area are not counted in GDP. They can only influence GDP negatively if the price of energy shoots up and if the general price index is pulled up noticeably as the consequence. From the discussion of the first part of this comment, this kind of effect is not likely in the era that the decoupling between the general price index and the price of energy is deeply entrenched. Then there is the increased government spending that will increase GDP. Considering all those factors mentioned above, we anticipate that the hurricane will reduce less than 1%, probably far less, from the growth rate of real GDP. At the time when real GDP is growing like 3.5% a year, this temporary reduction of the growth rate will not push the economy into a recession. When the recovery phase begins, all the rebuilding of lost properties and extra spending to repair the damaged houses will give GDP a strong push upward either from the component of residential fixed investment or from the component of final sales. Ironically the substantial damage to GDP after a sever disaster is rather caused by the psychological shock of the general public not directly affected by the disaster; if they spontaneously reduce their spending, then GDP will suffer a meaningful blow. In conclusion the effect of this hurricane will be the same as other major disasters, including the disaster of 911. Initially it may cause the growth rate of GDP to dip somewhat, not necessarily from the actual damage to the components of real GDP but due to the shock on the general public. Then GDP will bounce back strongly as the recovery phase starts. As time goes on, those negatives and positives from a disaster will average out and leave real GDP in a state that it has been before the disaster and continue its course as if there were no disaster at all. To properly gauge the effect of a disaster like hurricane Katrina, it is instructive to consider something like a real National Account that counts all the values of assets of the whole nation adjusted for speculative price movements on stocks and real estates, along side with real GDP. Then a sever disaster will depress the real National Account by a substantial amount. It will provide a proper picture of the hardship caused by the disaster that real GDP cannot gauge in full.