Business, Government, and the World Economy

1
Business, Government, and the World Economy

• Consumption

2
Aggregate Demand

• The amount that consumers, business and
  Government wants to purchase.
• Consumption
• Investment
• Government
• IS/LM Model joint determination of output and
  interest rates

3
Consumption

• Average or expected income of consumers
• Cost and availability of credit
• The decisions to save vs. spend

4
Determinants of Consumption

• Disposable Income
• Expected average Disposable Income
• Changes in Tax Rates
• Cost and Availability of Credit
• Demographic factors
• Expected Rate of Return on Investment Assets
• Changes in spendable income from asset price
  fluctuations (not part of DI)
• Consumer Attitudes

5
Empirical Evidence

• Consumption /Income is constant in long run
• Short run changes in consumption is not as
  closely correlated with income
• Cons based on average or expected income
• Attitudes shift frequently
• Cost of credit changes quickly
• Consumption is smoother in short run than income
  except during recessions

6
Empirical Evidence

• Changes in realized capital gains are not
  reflected in Disposable income, (similarly a
  reduction in mortgage rates could increase
  disposable income)
• Declines in available credit can have a larger
  influence on consumption than expected.

7
Variability of Consumption Spending

• 1990 recession consumption spending declined
  .2 over first 5 quarters
• 2001recession consumption spending increased by
  more than 3 (income growth, consumer sentiment
  both declined)

8
Quarterly Variability

• Changes in disposable income explains less than
  25 of changes in consumer spending
• Changes in money variables such as yield spread,
  money supply, stock prices, consumer credit also
  only 25
• The remainder of the fulctutaion is expalind by
  exogenous and random factors political shocks
  ec.

9
Long-term Variability

• About 75 of the one year change can be explained
  by disposable income, costs, availability of
  credit and asset prices
• For a 3 year change over 90 of the of the change
  can be explained by changes in disposable income
  etc.
• Managers need to be careful to not over respond
  to short term shifts.

10
Testing

• The variability on the past slides was quoted in
  the book lets test it using regression analysis
  (for a review) and data from the bea.

11
Regression Review

• Equation of a line Y a bX
• Graphing combinations of X and Y form a line.
• X is the independent variable and placed on the
  horizontal axis. Y the dependent variable and
  placed on the vertical axis (The value of Y
  depends upon X)
• a is the Y intercept and b the slope of the line.

12
Observations of X and Y variables

Y
X
13
Regression Estimates the line that best explains
the relationship between the variables

14
The Line is the one that minimizes the sum of
the squared residuals

15
Estimating the Regression

• The slope of the line is then equal to
• The Intercept is

16
Confidence in the ResultsR-Squared (R2)

• R2 will range up to one. It is the portion of
  the relationship explained by the regression
• R-Squared (R2) correlationYX2b2sx2/sY2
• Examples
• An R2 of one implies all the points are on the
  line
• An R2 of 0.5 would mean that half of the
  relationship is explained by the line.

17
Confidence in the ResultsT-statistic

• The t-statistic tells us whether or not we can
  reject the hypothesis that the variable is equal
  to zero.
• The higher the t-statistic the higher the
  confidence that we can reject the hypothesis that
  the slope is zero.
• If you cannot reject the hypothesis -- It implies
  that the dependent variable has no impact on the
  independent variable.

18
T-Statistic

• A Rule of Thumb
• The confidence levels are based upon the number
  of observations, but in general
• If you have a t-statistic above 2.0 you can
  reject the null hypothesis at the 95 level.
• (With 120 observations a t-statistic of 2.36
  allows rejection at the 99 level)

19
Standard Error

• Provides a measure of spread around each
  variable.
• Provides a confidence band similar to standard
  deviation)
• We can use standard error to estimate the T-
  Statistic (Assuming a normal distribution)
• T-StatinterceptA/SEA T-Statslope B/SEB

20
Quick Review

• Linear Regression - Provides line the best
  describes the relationship between two variables
• R2 - Portion of relationship explained by the
  estimated line
• T-Statistic - Confidence in the estimate of the
  variable (Is is statistically significant?)
• Standard Error - Confidence Interval

21
Quarter to Quarter Results
22
Yearly Changes
23
The Consumption Function

• Basic idea is trying to determine the
  relationship between consumption and the key
  inputs.
• Start with a simple model
• Assume that consumption is determined by
• Disposable income (DI)
• Marginal propensity to consume (MPC)
• C a (mpc)(DI)

24
Impact of consumption

• We said before that output in the economy equals
  income.
• C G I NX
• Assume no trade so output C G I NX

25
Planned Expenditure

• Substituting the consumption function into the
  equation for C produces
• a (mpc)Y G I
• Let this equal the amount of planned spending in
  the economy.
• mpc will generally be less than 1

26
Keynesian Cross

• Graph Planned expenditure on vertical axis and
  output (or income) on the horizontal.
• Let Y be the equilibrium level of spending where
  planned spending equals output (income)
• When output is less than Y
• Planned Spending gt Output (income)
• Inventories decline production increases
• When output is greater than y
• Planned Spending lt Output (Income)
• inventories increase, production decreases.

27
Old interpretation

• Given that the data supports that low income
  workers are dissaving and high income workers are
  saving
• Can it be said that
• The personal saving rate increases as income
  increases? (Keynes)
• Total GDP can be increased by transferring wealth
• Increases in income is accompanied by increases
  in saving that is not fully invested need to
  increase G relative to everything else

28
Assuming old Interpretation is correct

• The following outcomes which are not supported by
  data would occur if the old interpretation is
  correct
• Consumption would not rise as fast as income and
  savings would increase over time (as real income
  as increased)
• Consumption would be a function of only income
  not financial variables such as interest rates
  and credit
• The saving rate would decline in recession
• Saving rates would be higher for richer
  individuals

29
PIH

• Consumption is based on expected or permanent
  income, not just current income.

30
Permanent Income Hypothesis

• Those at the high end of the income scale have
  income above their long term expected income
  (permanent income) therefore they are saving
  (not because they have high permanent income)
• Those at the low end of the income scale have
  income below their long term expected income
  (permanent Income) therefore they borrow (not
  because they have low permanent income).

31
PIH (two periods)

• You can save a portion of your income. Let labor
  income Y then there is savings available in
  year 2 equal to
• (1r)(Y1-C1)
• The total amount available to spend in year 2 is
  then
• C2 Y2 (1r)(Y1-C1)

32
C2 Y2 (1r)(Y1-C1)

• Rearrange
• C2 (1r)(C1) Y2 (1r)(Y1)
• Divide by 1r
• C1 C2/(1r) Y1Y2/(1r)
• The PV of consumption must equal the PV of
  income -- in other words the key constraint on
  consumption is your lifetime income.

33
Intertemporal Budget Constraint

• Graph period 2 consumption on vertical axis (max
  value Y1(1r) Y2)
• Graph period 1 consumption on horizontal axis
  (Max value Y1Y2/(1r))
• Combine budget constraint with indifference
  curves (combinations of consumption with same
  utility)

34
An Increase in Income

• Assume that future income (Y2) increases by
  50,000.
• Assume that current income increases by 50,000
• In either case consumption increases in both
  periods
• Basic Perm Income model sets consumption equal in
  both periods.

35
PIH and mpc

• You still have the choice between saving and
  consuming the marginal propensity to consume
  still plays a key role.
• In both the Keynesian model and the intertemporal
  model an increase in permenant income will cause
  a large increase in current consumption

36
Precautionary Saving

• In reality future income is uncertain. The
  choice to save or consume is then in part based
  on precautionary saving (insurance against future
  uncertainty)
• This impacts the marginal propensity to consume.

37
Credit

• Given the intertemporal nature of the consumption
  decision, the amount of credit available and the
  cost of credit play key role in the decision to
  save or consume.

38
Current Consumption Theory

• Life Cycle Model of Saving and Consumption)
• People will attempt to borrow and save to keep
  the purchase of goods and services more stable
  than income.
• Everyone will act rationally to maximize their
  own self interest by
• Interpreting and Weighing Information
• Appropriately Balancing Evaluating Choices
• Making Informed Decisions

39
Life Cycle Model
Entrance to Workforce
Retirement
Age
40
Implications of Life Cycle ModelSaving Decisions

• Individuals understand the need to save for
  retirement and can estimate the amount they need
  to save.
• In other words consumers
• Understand the impact time has on the value of
  their money.
• Make informed decisions about their investment
  choices and actively respond to changes in the
  economic environment.
• Act in a manner that maximizes their investment
  income.
• Can accurately plan for a retirement age.

41
Life Cycle Implications

• Individuals attempt to smooth consumption
• If income drops due to short term layoff the
  expectation is that consumption would not
  decrease as much as income.
• If income drop is viewed as permanent
  consumption may drop by the same amount as income.

42
Extension

• Individuals at the high end of income scale -
  should have current income higher than their long
  term expected income - they should save
• Individuals at the low end of the income scale
  should have current income less than their long
  term expected income they should dissave
  (borrow)

43
Some real world dataIs PIH correct?

• Only 42 of workers have calculated how much they
  need for retirement. (EBRI 2006).
• 30 of US workers have not saved anything for
  retirement (EBRI 2006).
• Consumption patterns indicates that US workers
  experience an unexpected drop in standard of
  living after retirement. (Bernheim et. al 2001)