Kevin Clinton Winter 2005 Lecture notes 1 Macroeconomic data and history Facts to explain 1. Facts, theory, and policy In macroeconomics we deal with the big picture i.e. major aggregates in the economy. These include: levels of output, income, and employment; inflation (the rate of increase of the overall price level). Policy has objectives with respect to these: i.e. to increase output and employment, and to maintain a low rate of inflation. The relevant policy instruments are fiscal policy the overall tax and spending decisions of the government, and the net balance, but not the components) monetary policy in Canada the joint responsibility of the government and the Bank of Canada, but implemented autonomously by the central bank in brief, the instrument is short-term interest rate Theories, or models, are designed to explain the facts. Useful policy analysis and recommendations requires a sound framework of theory. Conversely, in this course we will focus on models that have direct application to practical policy problems. 2. Output and spending and income Circular flow GDP measures the aggregate value of output of goods and services in the economy. It also gauges spending, or aggregate demand, since each dollar of output corresponds to a dollar of sales (counting investment in inventory, whether desired or not). Likewise, it also indicates total income since each unit purchase generates an equal unit of factor payments wages, profits, etc. Related measures in the national accounts are GNP and GNE. Over time these variables provide much the same picture of the growth of the economy. All these aggregates are flow variables, measured over a given period of time (quarter, year, etc.). Stock variables, in contrast, are measured at a given point in time these include assets and debts. The change in a stock corresponds to a flow e.g. the net flow of inventory investment in a year is equal to the change in the stock of inventory measured at year-end. More generally, net investment adds to the capital stock, fixed or inventory.
Real versus money or nominal Unless otherwise noted, GDP is usually presented in real terms i.e. in constant-value dollars rather than current dollars. Effectively, this adjustment removes the impact of inflation, which is present in the money, or nominal, value of GDP (similar adjustments apply to other national accounts variables). To convert nominal GDP to real, we divide by a GDP price index or deflator. For convenience, we this index set to unity in the base year e.g. 1997 in Chart 1. (Pre-base year, because of inflation, the index is less than unity, so division by the deflator actually inflates nominal GDP.) Per capita GDP standard of living Per capita GDP provides a rough measure of average standards of living. For many purposes, the measure is good enough. For example, over time, the growth of per capita GDP in Canada does broadly reflect the expansion of the material goods and services we enjoy. But for other purposes e.g. for valid comparisons of standards of living across economies especially widely different economies you have to make adjustments. 2
3. Components Expenditure A famous identity puts total spending into 4 categories (the percentages give the approximate breakdown in Canada): consumption (60%) gross investment (17%) government (22%) net exports (1%): Y = C + I + G + (X M). Investment may be in fixed assets (buildings, machinery and equipment) working assets (inventory) intellectual assets (e.g. software development). Household outlays for residential construction are counted as fixed business investment in the national accounts. Other household spending is counted as consumption, even though an economic case can be made for treating consumer durables (e.g. autos, PCs) as well as some services (e.g. private education), as investments. Only spending on final output counts. Purchases of existing assets (e.g. used cars, existing houses) do not count (although the associated retail earnings and commissions do). Spending on intermediate goods is included only to the extent that it adds to the stock of inventory capital. To include intermediate goods used to produce final goods would mean double counting. Income GNP = GDP net factor payments abroad (or plus net foreign factor income) NNP = GNP depreciation. Depreciation of capital (capital consumption) reduces the net stock of capital, and reduces output available for direct enjoyment. At about 13% of GDP it is a large item with gross investment at 17%, net investment is only about 4% of GDP. national income = NNP indirect business taxes National income, on the earnings side, can be put into 4 categories: wages and salaries (72%) corporate profits (12%) non-incorporated business income (9%) net interest (7%). Thus, the labour share of national income is about 72%, the capital share 18%. 3
personal income = national income retained corporate profits net interest + government transfers to individuals social insurance contributions + personal interest income personal disposable income = personal income personal taxes 4
4. Output trends and cycles Chart 1 shows per capita GDP for Canada over a long period, and 2 post-ww2 subperiods. [Note: effective presentation of data is important. Be sure to understand why I use log GDP rather than the just constant-dollar level.] Chart 1 Canadian GDP growth 1926-2002 per capita constant 1997 $ natural logarithm trend = 0.024t + const R 2 = 0.96 1925 1945 1965 1985 2005 Post-1973 deceleration tre n d = 0.0 1 6 t + co n s t R 2 = 0.9 5 1 9 7 0 1 9 8 0 1 9 9 0 2 0 0 0 The chart shows some features important to our well-being and pervasive across countries that we should explain: tendency towards trend with high employment o post-ww2 deviations from trend have not exceeded 5% in size, or 5 years in duration o even so, the 1990s were fairly dismal for Canada a below-trend decade long-run expansion (upper quadrant) o the per capita trend line shows an annual increase of about 2 1/2 % o population growth has been steady at about 1 ½% so the trend growth was about 4% 5
post-1973 deceleration (lower quadrant) o the per capita trend down to about 1 ½% o in contrast, the 1950s and 60s saw exceptionally high growth per capita over 2 ½% business cycles o GDP movements around trend are not random, but serially correlated a below-trend (above-trend) year is likely to be followed by more of the same i.e. cycles o business cycles have recession, trough, recovery and peak phases (where are we now?) 1930s Great Depression o analysis of this catastrophe is to macroeconomics as pathology is to medicine Trend GDP and potential or full-capacity GDP There is hardly a more important topic in economics, since it determines standards of living. Growth theory relates long-run expansion to the supply side factor inputs the labour force, education and training, capital, technology, etc. These cumulate incrementally over time. Growth in demand cannot of itself produce expanding output in the long run. Spending in excess of productive capacity eventually results just in inflation. The long-run concept of monetary neutrality holds that movements in nominal variables like spending do not, in the long run, affect real variables. Since the GDP trend line corresponds to situations in which the unemployment rate is usually about average, trend GDP can be regarded as a rough indicator of potential GDP (there are better measures one is published by the BofC which take explicit account of those factor inputs). Cycles If potential output is given by supplies of inputs, which do not change much in the short run, business cycles must primarily be demand-driven. For better or for worse, monetary and fiscal policies have a clear role in the determination of short-run aggregate demand. At the trough of recessions demand is less than potential supply (negative output gap), whereas at the top of the recovery demand is typically above potential (positive output gap). NB In practical terms, the short run in macroeconomics may mean anything from 1 month to 5 years. Typically in this course you can take it to mean a year; we will usually look at annual data. 6
5. Unemployment Cycles in output are reflected in the labour market. For example, in Chart 2 you can see the sharp increases in the unemployment rate during the recession of the early 1980s and 1990s. However, in Canada UE rate movements are an awkward economic indicator major changes to unemployment insurance and demographics, etc. have altered the economic meaning of the unemployment rate e.g. in the 1970s strong growth in output and employment were accompanied by a rising UE rate. Cross-country comparisons are also tricky e.g. the published Canadian UE rate is above the US rate, despite much stronger recent employment growth here (a greater percentage of the population is in the workforce in Canada). Chart 2 14 12 10 8 6 4 2 0 1950 1960 1970 1980 1990 2000 2010 7
Okun s Law In the 1960s Arthur Okun found, in the US, an empirical correlation between change in output and changes in the unemployment rate that has survived surprisingly well over the years even in Canada, despite the structural changes in the labour market. Chart 3 Okun s Law for Canada Increases in GDP and unemployment 1973-2002 (percent) 8 6 y = -1.66x + 3.22 R 2 = 0.64 4 2 0-2.00-1.00 0.00 1.00 2.00 3.00 4.00-2 -4 The estimated line suggests that a one percent increase in the unemployment rate from one year to the next is associated with a decrease in output of about 1.7% (The original Okun s Law had 2.5%). The line also suggests that with no change in unemployment, output normally increases by about 3% p.a. an indicator of the rate of growth of potential output over the period. 8
6. Inflation and interest rates CPI a measure of the cost of living for typical households fixed weights, updated every 10 years to reflect the shifting consumer basket almost never revised (by virtue of the fixed weights) published monthly, about 3 weeks into the next month may be affected in the short run by special factors (volatile components, e.g. fuel, fresh food; or changes to indirect taxes and mortgage interest rates) The most common measure of inflation is the percentage annual increase in the CPI. Used for the BofC s 1-3% inflation target. The BofC calculates a core CPI (CPIX) inflation rate excluding components dominated by special factors. Because of its lower volatility, the CPIX is a better guide than the CPI itself of underlying changes in the CPI. GDP deflator calculated as nominal GDP/real GDP a comprehensive measure of the price level of final goods & services, including o consumer items o investment o exports o government (services and equipment, etc.) current implicit weights, which vary with spending patterns revised regularly, with revisions to the national accounts estimates published quarterly, 2 months into the next quarter may be significantly affected by changes in raw material prices, including forest product, minerals and energy prices, since Canada is a large net exporter of such materials Although year by year the 2 main price level measures show visible differences, over long periods of time they have shown similar average rates of inflation. 9
CPI inflation history Chart 4 3-month t-bill rate and year/year increase in CPI 1952-2002 % 20 15 10 5 0 1950 1960 1970 1980 1990 2000 2010-5 Features to explain: stabilization of inflation at about 2% since 1992 high rates of inflation in the 1970s and 80s Interest rate nominal and real The movements of the t-bill rate are fairly representative of short-term interest rates, although other yields are higher, reflecting risk premiums. There is a clear, but far from perfect, correlation between the interest rate and inflation. The rate we observe the nominal interest rate contains a premium for expected inflation. This premium protects financial asset values from inflation-erosion; borrowers are willing to pay it because they benefit (at least nominally) from inflation. Large lasting changes in the inflation rate are followed sooner or later by changes in interest rates. The Fisher Equation states that, in equilibrium, the nominal interest rate is equal to the real rate plus the (unobserved) expected rate of inflation. i = r + Eπ This is another example of the concept of monetary neutrality. We can calculate, conversely, an approximation for the actual real interest rate by subtracting the actual rate of inflation from the nominal interest rate (Chart 5). (With a steady inflation rate, people would come to expect it). 10
Chart 5 Real t-bill rate 10 8 6 4 2 0-21950 1960 1970 1980 1990 2000 2010-4 -6 While the average real 3-month rate has been 2-3%, for some extended periods it has been significantly below (1970s) or above (1980s) that range. These call for explanation in terms of monetary policy (which sets the short-term interest rate) and expectations of inflation, which may turn out to be inaccurate. In some cases, the influence of monetary policy is clear. For example, the BofC deliberately set high real interest rates in the early 1990s to reduce inflation from 5% to a target eventually set at 2%. And the BofC set low rates in 2002, in the face of weaker demand from the US, to forestall unwanted disinflation. On the other hand, growing confidence in low inflation produced the general down-trend in interest rates in the 1990s. 7. Conclusion Canadian economic history turns up plenty of important questions with respect to longrun and short-run movements of output, employment, inflation and interest rates. Once we have a framework to explain these variables, we can talk about fiscal and monetary policy in an informed way. 11