Economic Fluctuations
in Growth-Trend Perspective

A Growth-Trend Conceptual Framework
For Empirical Analysis of Macroeconomic Performance

The words we use influence the way we think. An essential attribute of a real science is its use of precise, consistent, and analytically significant defintions. Traditional uses of the terms "recession," "depression" and "recovery" do not meet this requirement. This conceptual framework provides a more scientific basis for analyzing and evaluating economic performance.

Growth-Trend Conceptual Diagram

The letter-designated aspects of this chart illustrate the following conceptual definitions.)

Statistically Measurable Concepts

[A,B & C] are in "real" dollars of constant purchasing power (which exclude inflation effects), and are on a ratio scale -- where a straight line has a constant growth rate and equal vertical distances have equal percentage changes. [F,G & H] are % of CGDP. [E,R & Q] are growth rates. These are different essential ways of quantitatively analyzing and evaluating the basic GDP measure of economic performance.

[A] Capacity GDP (CGDP) -- also called Potential GDP or Full-Employment GDP. This is an estimate of the economy's optimum output with effective full employment of its labor force. See more on the conceptual definition and statistical estimation of Capacity GDP and Full Employment.

[B] Actual GDP (AGDP). This series is preferably smoothed to approximate the underlying trend.

[C] GDP Gap (CGDP - AGDP). The dollar amount of GDP lost from depression.

[D] Growth Rate Protractor. Used with a clear parallel-line overlay, this invaluable analytical tool makes it possible to measure the growth rate between any two points, or the trend growth rate during any particular period.

[E] Actual Growth Rate (AGR) (right scale) -- preferably smoothed, to approximate the underlying current trend. Its two functional components are:

[F] Operating Rate (OR) -- the ratio of Actual GDP to Capacity GDP. This corresponds to the traditional manufacturing industry term,"Capacity Utilization Rate," which, for individual plants, is also often abbreviated to "Operating Rate."

[G] OR Gap (100% minus OR). This corresponds to the unemployment rate -- according to the "Okun's Law" rule-of-thumb, a 1% change in unemployment rate corresponds to a 2 1/2 - 3% change in the OR Gap.

General Descriptive Terms For Economic Performance

The growth-trend conceptual framework, as illustrated in the diagram makes it possible to give precise conceptual definitions and empirical dimensions to traditional business fluctuation terms. An diagram illustrating an analogy to a hole in a road on a hill may also prove helpful.

[H] Depression (below Capacity GDP) -- When actual GDP is depressed, and unemployment is correspondingly excessive. The severity and cost of depressions are precisely measured by the OR Gap (%) and GDP Gap ($ amount). (The traditional idea that a "depression" is a "severe recession" is too imprecise, both conceptually and empirically, for a real economic science!)

Since, in the past, the economy has sometimes gone for long periods of more or less continuous fluctuation without ever closely approaching Capacity GDP, it is also appropriate to think in terms of several briefer but more severe depressions within a longer span of sub-capacity OR -- like craters on the moon or potholes in a road -- as illustrated in the diagram, and during 1930-42 and 1973-97.

[J] Recession. The Operating Rate is receding -- even if the GDP growth rate is above zero):

[K] Contraction -- as "officially" designated by the prestigious National Bureau of Economic Research (NBER). This refers generally to a period of absolute negative economic growth, as distinguished from the relative decline of a growth-trend recession. (For a discussion of other misleading analytical implications of the "static" (absolute-based) NBER conceptual framework, see "business cycle".

[L] Stagnation -- when there is no significant change in the OR and the OR Gap. LH is high-level stagnation, KL, low-level stagnation.

[M] Recovery -- an increase in the Operating Rate (not just the growth rate).

Evaluating and Managing the Recovery Growth Track (RGT)

Almost everybody today realizes that the Federal Reserve is able to, and actually does, manage the economy's overall growth and unemployment rates with a fair degree of precision. In that perspective, it is important to analyze past and potential Recovery Growth Tracks [M,N,O,P], and to recognize explicitly that every RGT has two distinct growth-rate components:

The Unemployment/growth-rate/inflation relationship.  For effective empirical research on this relationship it is important to separate the RGR from the CGR, and to focus on the interaction between the OR Gap and the RGR. This is a far more scientific and productive approach than the crude Phillips-Curve/NAIRU approach, which relates inflation only to the unemployment rate (which tends to lag the OR Gap by several months), and ignores the functionally-crucial RGR. Moreover, it is only the RGR component which can be managed effectively by Federal Reserve monetary policy.

Achieving the fastest non-inflationary recovery

This requires a "soft-landing" (asymptotic) RGR -- initially fast, when the economy has many unused resources, then gradually slowing as it approaches Capacity GDP, to allow our free-market economy to make the necessary structural readjustments and capital investments -- including labor force (re)education and training -- that are needed to reach a full Capacity OR. The growth-trend conceptual diagram illustrates three types of RGR.

Past history suggests that in a typical recovery the fastest non-inflationary RGT is produced by an RGR equal to approximately 1/2 of the OR Gap. For example, if the OR Gap is about 12% (and the corresponding unemployment rate about 7%), the optimum initial RGR would be about 6% and the initial AGR about 8-9%.

However, if the recession-inducing inflation is caused by factors other than "free-market" supply/demand relationships -- i.e., by monopolistic price manipulations such as the OPEC cartel's 1970s "oil tax" price increases, or by government taxes or regulations which directly impact business costs and/or prices, the formula growth rate must also be accompanied by specific government interventions to counteract these non-market (and "inflation factors.

Moreover, sophisticated selection of the most appropriate specific formula should take account of several key "environmental factors" which make it more difficult for the Federal Reserve to quickly achieve a capacity OR without inflation unless the Feds monetary policy is accompanied by coordinated assistance from other government policies:

Duration of preceding depression.  The longer the preceding depression (regardless of its depth), the more damage it will have done to the economy's basic productive capacity - - its physical buildings and equipment, properly skilled labor force at appropriate locations, experienced and well-organized management, actual application of available new technology, etc. -- and the longer it will take to achieve the necessary rebuilding and structural readjustments.

Preceding inflation experience and "inflation expectations."  The two bouts of world-wide inflation after the OPEC cartel quadrupled oil prices in 1973-74 and levied another huge "oil tax" in 1979-80 caused a severe inflation phobia. In this environment, economically appropriate rapid growth tends to be considered inflationary and causes central bankers to place highest policy priority on preventing inflation, while everyone else (particularly bondholders) try to make sure that theirincomes stay ahead of anticipated inflation.

Political environment.  In a political environment dominated by free-market laissez-faire philosophy, where the inherent cyclical tendencies of capitalist economy (including "boom" financial excesses during recovery) are allowed to play out unmitigated by government intervention, achieving a full-employment OR without inflation is very difficult. It is easier where government accepts responsibility for managing the economy for that objective.

For example, in 1940, after 10 years of the 1930s Great Depression, with 25% unemployment in 1933, many economists despairing of ever again reaching 1929's 3% unemployment rate because workers had "lost their job skills and work discipline". But after Hitler's invasion of Europe in May 1940, the political environment became strikingly different as America began rearming for our active participation in World War II. When the government put maximum output ahead of ideology and systematically coordinated all available economic policies, GDP shot up far above its long-run capacity growth-trend, and unemployment plummeted from 15% in 1940 to below 2% in 1943. Formerly unemployed Southern sharecroppers, housewives and even "disabled" people suddenly became skilled workers in Detroit and California airplane factories. All with quite limited inflation! Nowhere was the old adage more relevant: "Where there's a will there's a way"!

Public confidence (business and consumer).  The investment decisions which are so vital to economic recovery and which usually involve borrowing, whether by businesses for plant and equipment or by households for houses and cars, are best made with a relatively long economic time-horizon. Their volume will be greater and earlier when people have confidence in their economic future.

Thus, the optimum soft-landing recovery formula at any particular time depends very much on the degree to which the federal government explicitly accepts responsibility for managing the economy, has credible policy tools for doing so, and systematically coordinates them so as to promote balanced growth and optimum output -- including responsible use of all potentially available anti-inflation policy tools.

Conceptual Definitions and Statistical Estimates

Conceptual Definitions

"Capacity GDP" and "potential GDP" are often used synonymously. But "capacity" has the advantage that it is also used in the same sense for individual manufacturing plants and industries, where it is the denominator of the key analytical ratio, "capacity utilization rate," (often shortened to "operating rate," as it is in this conceptual diagram [F]).

And just as Manufacturing plants often operate above 100% of their normal "rated" capacity for brief periods by postponing maintenance, working extra shifts or using high-cost "reserve" equipment, Actual GDP usually exceeds 100% of normal capacity during wartime. Thus, it is useful to reserve the term "potential GDP" for that higher limit.

"Full-Employment," like "Capacity GDP," can best be defined in practical but explicit terms: an economic environment in which everyone who needs and wants to work for pay can with reasonable effort and minimum seek-time find a job which reasonably fits his/her education and skills, and pays a corresponding income.

(The liberal explicit proviso of "a living wage" for even the lowest-wage workers is probably unnecessary because a sustained economic environment of stable full-employment growth can only be achieved by a systematically coordinated package of economic policies which also maintains a sustainable structural economic balance, and such a proviso is impossible to define explicitly because in practice it depends on the number of workers and dependents in a household and how efficiently their income is spent.) In a dynamic, flexible economy like ours, some workers will be always be "between jobs," even in a generally full-employment environment. But extended involuntary unemployment is one of the greatest affronts to individual human dignity, family values and youth welfare. In an effective democracy, people should be not forced into long periods of unemployment or involuntary early retirement by mismanaged economic policy. Moreover, experience around the world continues to demonstrate that high unemployment is a major threat to social and political stability, fiscal solvency, and social insurance programs.

Labor market effects.  The other side of the full-employment coin, of course, is that some employers, in some locations or industries will, have difficulty finding additional employees with the optimum experience and skills they want at the wage or salary they would prefer to pay. Thus, there will be continuous incentive for both employers and governments to assist workers in obtaining the education and skills they need for optimum participation in the economy, and to develop a more efficient "labor market" for matching workers and jobs.

"The inflation threat." ........(incomplete)

The conceptually and morally malignant concept of a "natural" unemployment rate.  After 10 years of the 1930s Great Depression, many economists despaired of ever regaining full employment. Similarly, after 30 years of seemingly intractable inflation (due only partly to OPEC's 1973 and 1979 "oil tax" increases), many economists despaired of ever again achieving full employment without inflation. They argued that there was a "natural" rate of unemployment that is consistent with stable prices (the Phillips-Curve-based NAIRU -- non-Accelerating-Inflation rate of Unemployment) and called the resulting output "Full Employment GDP." But just as the depression-era pessimism was discredited by WW II experience, the inflation-era pessimism was discredited by the joint reduction of both unemployment and inflation during 1992-97. Neither is any more "natural" than high rates of poverty and economic inequality, political corruption, polluted rivers and oceans, depleted fish stocks and poorly-educated labor force; all are the results of bad public policies which need to be changed.

Statistical Estimates

For really scientific analysis of economic growth and fluctuations, statistical estimates of Capacity GDP must be based the economy's two basic functional components, labor force and productivity (GDP per hour of work).

Making a credible estimate of the long-run trend of Capacity GDP is much more difficult during a long period of depressed operating rates. But credible estimates both the labor force and productivity components can be made by adjusting the actual current values for the effects of sub-capacity operating rates. And since the U.S. economy is nearer to real full employment in mid-1997 than at any time since 1969, it is extremely important that this year be used as a benchmark for making new estimates of the long-run trend of Capacity GDP.

Smoothing Actual GDP

Erratic reported GDP growth rates which fluctuate wildly from 6% to 1% to 4 1/2% in successive quarters are analytically confusing and make it difficult to formulate sound economic policy. There are several ways to reduce this statistical "noise."

By better Fed monetary Policy.  Part of this "noise" is due to imprecise Fed monetary policy. In the context of more systematic coordination of all available <> economic policy tools, the Federal Reserve could significantly stabilize actual GDP growth by more precisely monitoring the basic trend of the economy's <> Money Demand Ratio (M1/GDP) and managing money growth accordingly.

But much of the apparent erratic performance is functionally unimportant "statistical noise" which makes it difficult to determine the underlying economic growth-trend. To provide financial markets, economic forecasters, business planners and government policy makers with a more useful analytical tool the Bureau of Economic Analysis should estimate and publish a "Trend GDP" series which eliminates inventory fluctuations that don't correspond with the basic trend of GDP growth, the effect of computerized seasonal adjustment formulas which don't accurately reflect current "seasonal" influences, and other such influences. This series would serve analytical purposes similar to the Labor Department's "core" inflation series that excludes volatile food and energy prices

Inventory fluctuations.  When the growth rate of final demand falls abruptly (often due mainly to imprecise money-growth policy), inventory in the production and distribution "pipeline," which tends to be geared to the previous rate of final demand, tends to back up -- particularly in terms of closely-watched inventory/sales RATIOS. This causes business firms to cut back production rates, which leaves pipeline inventory inadequate for the next spurt of final demand.

Focusing on Final Sales rather than total GDP, and using an average of GDP and final sales are sometimes useful for estimating the underlying growth-trend. But both are inadequate because official Final Sales data are also adversely affected by transitory factors such as automobile price "rebates," other business "promotions" and weather factors. To the extent that consumer purchases of cars and other major durables are basically limited by current income and the current burden of instalment debt repayments, the effect of special sales promotions is mainly to borrow from future sales. So, data on consumer spending minus increase in instalment debt increase is also highly useful analytically.

But a more sophisticated approach is to recognize that business inventory fluctuations are strongly influenced by corresponding fluctuations in consumer inventories, which in turn are closely related to fluctuations in consumer instalment borrowing. The basic trend of consumer durable goods inventory stocks can be derived from the Bureau of Economic Analysis Wealth Data Tape and related to consumer spending, income and instalment borrowing, as basis for adjusting GDP.

Poor seasonal adjustments.  This is a major cause of statistical noise when actual "seasonal" factors, particularly weather, do not correspond with the historical-based averages. With the power of modern computers it should be possible to take specific account of the effects of "unseasonal" events (heat and cold waves, rainfall, floods, changes in auto model changeovers, number of days and holidays in a survey period etc.) which have significant effect on economic activity. A major cause of bad GDP seasonal adjustments GDP in the 1970s was the fact that the OPEC "oil tax" caused huge changes in economic patterns, while the BEA continued to use a historical- based computer adjustments. Appropriate "seasonal" adjustments should take account of all the known factors causing deviations from the current basic trend of the data.

Conceptual Problems with the NBER "Business Cycle" Model

Although popularly called "recessions" NBER-designated "contractions" should actually be considered severe recessions because a year of even zero AGR tends to cause about 1% more unemployment, as well as more than 2 1/2% loss of GDP, and over $70 billion increase in federal deficit.

Many years ago, when the NBER realized the problem caused by this popular mis-designation, they started using the term "growth recession" for a decline in the AGR. But since a "soft-landing" (asymptotic) decline in the AGR is actually beneficial during an optimum recovery the "growth recession" concept also tended to give an analytically wrong impression and wasn't very useful.

Another misconception stemming from the NBER "static" (absolute-base) "business cycle" conceptual framework is the purported "lead" of the so-called "Leading Indicators", which is largely conceptual rather than economically functional. Since a growth-trend recession (negative RGR) necessarily precedes (leads) an NBER contraction (negative AGR), and since most of the "leading" indicators are functionally related to the AGR, the durations of their highly variable "leads" depend mainly on the time-lag between the onset of a growth-trend recession and an NBER contraction -- the longer this lag, the longer the indicator "lead."

In the typical "cyclical" upturn, which tends to be relatively abrupt and clearcut, the timing difference between the beginning of the growth-trend recovery and the NBER-designated "expansion" is relatively insignificant. However, since the basic concept of "business cycle" is a myth, there have been several significant periods (1971, 1975-76, 1990-91) when the economy had a relatively long period of continued growth-trend recession or low level stagnation after the NBER-designated beginning of "expansion" before beginning recovery of the Operating Rate

Last revised: April 30, 1998
To IEA home page