Long-term economic growth is linear

In this blog, we have discussed already [here and here] the incompatibility of real GDP data caused by the change in definition of the GDP deflator, dGDP, (in the USA - in 1977).  Figure 1 shows details of the deviation between the dGDP and the consumer price index, CPI, as expressed by the cumulative inflation rates. One can see that the CPI inflation rate is approximately equal to the rate of the GDP deflator change multiplied by a factor of 1.22 since 1978. This allows recovering the dGDP time series back in time using the strong statistical link between CPI and dGDP (1.22dGDP = CPI) , as shown in Figure 1.

This observation naturally leads to the assumption that real GDP in the United States is biased by the change in definition of the GDP deflator. (According to “Concepts and Methods of the U.S. NIPA” the growth rate of real GDP is the growth rate of nominal GDP reduced by the overall change in prices as expressed by the GDP deflator or the economy-wide price index.)  The Bureau of Economic Analysis warns economists that the real GDP time series is incompatible over time. What is the consequence of this bias to the long-term economic trends? One can correct real GDP time series for the definitional change and estimate the change in trend.  Figure 2 shows real GDP and real GDP per capita in the USA from 1929 to 2013. The latter time series has rather a linear trend since 1929 with Rsq. =0.97. The real GDP series deviates from the long term exponential trend since 2000 – the year then the rate of population growth fell below 1% per year.

In Figure 3, we correct real GDP per capita for the difference between CPI and dGDP after 1977 and compare the original and corrected time series. One can see that the corrected curve has Rsq.=0.99 and does not deviate from the long-term trend. Currently, the corrected growth rate is slightly below the long-term trend. (Obviously, we could correct the period before 1977 and obtain the same statistical result.  Statistically, the Solow model (constant returns to scale) implying the exponential GDP per capita growth is wrong.

Figure 1.  Cumulative rates of CPI and dGDP inflation, original and scaled by a factor of 1.2.

Figure 2. Real GDP and real GDP per capita in the USA from 1929 to 2013. The latter time series has rather a linear trend since 1929. The real GDP series deviates from exponential trend since 2000 – the year then the rate of population growth fell below 1% per year.

Figure 3. The real GDP per capita time series corrected for the difference between CPI and dGDP since 1978. Linear trend is obvious in the corrected time series. Currently, the growth rate is slightly below the long-term trend.


A few steps down to destroy WTO

These days sanctions and retaliation  is a hot topic. The first round is over and we will likely observe escalation well supported by political rhetoric.  In the long run, the system of world trade falls with acceleration.  The problem is not the current ping-pong with billions lost. The root problem is the political coup de grace to the overall trust and confidence between business agents.  It is not clear how to behave and calculate risks if the level of loss is out of control.
Dark ages knock the door - claims in WTO will increase in number and volume. It is not excluded that WTO will be paralyzed in a few months.


Will EU sanctions produce recession in ... EU?

When calculating the outcome of sanctions political leaders focus on some simple figures of immediate loss for both sides. The problem is that they ignore real economic forces underlying the reaction of complex economies to any major disturbance.  The first response of the country imposing sanctions might seem minor when expressed in Euro, but it makes not only immediate harm but changes business risks throughout the whole economy. The new profile of economic risks affect financial system by increasing risk premium and slowing  business. The whole economic system, especially in the years of slow recovery, might sink into another recession period before reaching a stable growth path. Political elites feel this possibility, which is dangerous for them in terms of troubles for elections, and likely react wrongly by new sanctions mainly supporting the emotional side of punishment. A new round of sanction will likely increase the level of economic and financial risks exponentially. The only benefit is that the country under sanctions will suffer more.
As in boxing, the heavy-weight winner is always in bruises and own blood. But who cares when brain is damaged


Labor force participation rate will fall to 59%

A year ago we presented a description of secular fall in the labor force participation rate, LFPR, measured by the Bureau of Labor Statistics. The LFPR (the portion of people in labor force) for the working age population (16 years of age and over) has been on a long-term decline since 1995. We predicted the fall down to 59% by 2025. Here we revisit this projection and find that our forecast was correct – the rate has decreased by 0.6% (from 64.4% to 62.8%). This is a dramatic drop considering the level of labor force of 155 million.
Following the Kondratiev wave approach (the Russian economist Kondratiev introduced long-period (50 to 60 years) waves in economic evolution – see Figure 1) we interpolated the observed LFPR curve by a sinus function with a period of ~70 years. We added 11 LFPR readings published since July 2013 and show the updated curve in Figure 2. New data fit the predicted curve.  The trough of the model function is expected in 2030 and the bottom rate is 58.5%.


Figure 1. The Kondratiev wave

Figure 2. The actual LFPR curve (red) and that predicted by sinus function with a period of ~70 years.


Devastating depopulation of Ukraine

A year ago I presented a graph illustrating the evolution population in Russia and Japan. This post was called Catastrophic depopulation in Russia and Japan. All data were borrowed from the World Bank population projections given for all countries through 2050.
Definitely, the current events in Ukraine will affect the evolution of population. The World Bank expects 79% of the 2010 level in 2050, as Figure 1 depicts.  Twenty one per cent of population will be lost.

Figure 1. Depopulation of Russia, Japan, and Ukraine. All curves are normalized to their respective values in 2010.  Ukraine will reach 79% by 2050


Personal and national income collide

Income inequality is a hot topic for professional economists and lay public. Piketty’s book Capital in the Twenty-First Century (2013) attracts common attention and discusses income distribution between labor and capital. The root concern is related to increasing share of capital income. We made some comments on this topic showing that capital does not eat from the part of labor income but converts corporate income into personal income. Piketty projects some further growth in the proportion of capital income.

Here we present an extremely simple observation which bans any further growth in the capital ‘s share of income. Figure 1 displays the evolution of national income (NI), i.e. the sum of labor and capital income, and personal income (PI), both reported by the Bureau of Economic Analysis.  In the 1970s, the difference was 10% and then stared to decrease. This is the period which Piketty highlights as the era of capital income, i.e. all increase in the share of personal income was appropriated by capital.  Since 2011, there is no room for further growth in the share of capital income – all national income is distributed as personal income. There is no other source of income, except may be decrease in consumption of fixed capital (CFC). There is nothing to share any more.

Figure 1. Evolution of national income (NI) and personal income (PI) both normalized to Gross Domestic Product. Currently, they are almost identical. 


We expect the rate of unemployment at the level of 3% in 2015

In the USA, the rate of unemployment in May 2014 is 6.3%. Two years ago, we foresaw the rate to fall down to 6% [±0.4%] in the fourth quarter of 2013 or in the first quarter of 2014. According to our model, this dramatic fall from the level observed in 2012 is driven by the change in labor force and inflation. We foresee the rate of unemployment to fall down to 3% [±0.4%] in the beginning of 2015. In this post, we discuss the bottom rate of unemployment. The USA will meet a new situation with the smallest rate of unemployment since the late 1960s and low inflation. The performance of our original model is exiting.

In 2006, we developed three individual empirical relationships between the rate of unemployment, u(t), price inflation, p(t), and the change rate of labour force, LF(t), in the United States. We also revealed a general relationship balancing all three variables. Since measurement (including definition) errors in all three variables are independent it may so happen that they cancel each other (destructive interference) and the general relationship might have better statistical properties than the individual ones. For the USA, the best fit model for annual estimates was a follows:

u(t) = p(t-2.5) + 2.5dLF(t-5)/dtLF(t-5) + 0.0585   (1)

where inflation (CPI) leads unemployment by 2.5 years (30 months) and the change in labor force leads by 5 years (60 months). We have already posted on the performance of this model several times.

For the model in this post, we use monthly estimates of the headline CPI, u, and labor force, all reported by the US Bureau of Labor Statistics. The time lags are the same as in (1) but coefficients are different since we use month to month-a-year-ago rates of growth. We have also allowed for changing inflation coefficient. The best fit models for the period after 1978 are as follows:

u(t) = 0.63p(t-2.5) + 2.0dLF(t-5)/dtLF(t-5) + 0.07; between 1978 and 2003

u(t) = 0.90p(t-2.5) + 4.0dLF(t-5)/dtLF(t-5) + 0.30; after 2003

There is a structural break in 2003 which is needed to fit the predictions and observations in Figure 1. (This break is purely artificial because it was induced by new statistics of labor force and inflation introduced in 2003 by the Bureau of Labor Statistics. The same effect was in 1978, when main definitions of labor force, unemployment, and inflation were revised.) Due to strong fluctuations in monthly estimates of labor force and CPI we have to smooth the predicted curve with MA(24). As a result, the prediction horizon decreases from 30 to 18 months.

Figure 1 depicts the predicted and observed rate of unemployment since the beginning of the 1960s. Figure 2 depicts the observed and predicted rate of unemployment since 2000, including a forecast for the next 18 months. The model showed that the unemployment rate will fall to 3.0 % in the beginning of 2015. For 119 observations since 2003, the modelling error is 0.4% with the precision of unemployment rate measurement of 0.2% (Census Bureau estimates in Technical Paper 66). Hence, one may expect 3.0% [±0.4%].

 Figure 1. Observed and predicted rate of unemployment in the USA. 

Figures 2. The predicted rate of unemployment. We expect this rate to fall down to 3.0%  [±0.4%] in the beginning 2015.