Intermediate Macroeconomics

Transcription

1 Intermediate Macroeconomics Lecture 9 - Government Expenditure & Taxes Zsófia L. Bárány Sciences Po 2011 November 9

2 Data on government expenditure government expenditure is the dollar amount spent at all levels of government this can be in the form of: purchases of goods and services transfer payments (amounts given to households and businesses) interest payments to understand the magnitude, we compare government spending to GDP as a general trend government spending over time has increased as a fraction of the GDP there is a wide variation across countries in the level

3 Total government spending

4 Breakdown of government purchases

5 Breakdown of transfer payments

6 Breakdown of social security spending

7 Government spending across countries

8 The government s budget constraint I. like households, governments have different sources of income and different ways to spend it the budget constraint states that the amount of revenue has to equal the amount of spending G }{{} t + V }{{} t government purchases transfers = T t }{{} taxes + M t M t 1 P t }{{} revenue from printing money for now we will ignore the possibility of the government taking loans and repaying them (B g t = 0) revenue from printing money typically small assume that M t = M t 1, no money printing G t + V t = T t real purchases + real transfers = real taxes

9 Public production assume that the gov uses its purchases of real goods and services to provide services to households and businesses free of charge in most countries public services include: national defense, law enforcement, police & fire protection, elementary, secondary schooling, higher education (partly), health services, roads, parks, lighting, etc one option would be to model this as the output of the government s production function, in this case the government would have its own stock of capital, employees, etc to simplify: assume that the government buys final goods and services to the private sector C t + I t + G t is the total demand for final goods and services

10 Public services - what is their use? possibility 1: they give utility to the households examples: parks, libraries, school lunch programs, health care, public transport, the entertaining parts of the space program... substitute for private consumption - if a student receives the school lunch, they don t have to buy their own possibility 2: they are inputs to private production examples: provision and enforcement of laws and contracts, gov sponsored R&D programs, fire and police services, regulatory activities, the technologically valuable parts of the space program... substitute for inputs in the productions function (police for private guards), or increase their marginal productivity (infrastructure) hypothetical case: gov purchases have no effect on utility and production this basically assumes that the gov buys goods and services and then throws them away

11 The household budget constraint consumption + real saving = real income C t + 1 P B t + K t = w t P L t + r t 1 ( Bt 1 P ) + K t 1 + V t T t Extend this to the multi-year budget constraint: C 1 + C 2 C 1+r (1+r 1 )(1+r 2 ) +... = (1 + r 0)( B 0 P + K 0) w 1 P L 1 + w2 P L 2 1+r 1 + w 3 P L 3 (1+r 1 )(1+r 2 ) +... V 1 T 1 + V 2 T 2 1+r 1 + V 3 T 3 (1+r 1 )(1+r 2 ) +... The last part is the present value of real transfers net of real taxes.

12 Permanent change in gov purchases - effect on the household Look at the effect of a one unit rise in government purchases in every year: G t + V t = T t G t = T t V t = (V t T t ) a one unit increase in government purchases a one unit decrease in the value of real transfer net of real taxes a one unit decrease in the household s real disposable income if this is a permanent change and the labour supply is fixed from our previous analysis: the propensity to consume out of a permanent change in income is close to one a roughly one unit decrease in each year s consumption, C t

13 Permanent changes in gov purchases - effect on production I. Y t = A t F (κ t K t, L t ) capital demand is determined by the profit-maximisation of firms government purchases do not affect the marginal productivity of capital capital demand curve unaffected capital supply determined by the trade-off between depreciation and rental rates also unaffected by government purchases market clearing rental price R P and capital services (κk) do not change A unchanged and L fixed Y unchanged

14 Permanent changes in gov purchases - effect on production II. similarly government purchases do not affect the marginal productivity of labour labour demand curve, L d, does not change labour supply, L s fixed at L market clearing real wage, w P, unaffected Conclusions so far: a permanent change in government spending does not change κk does not change R P and w P does not change r = R P δ(κ) does not change Y

15 Permanent changes in gov purchases - effect on consumption r determines the intertemporal substitution since r does not change, intertemporal substitution does not change the labour-leisure choice is determined by the real wage rate, which does not change we also assumed that the labour supply (and hence the leisure) is fixed income effect is in play, the real disposable income drops by 1 unit in each period consumption decreases by about one unit in each period

16 Summary of the effects of permanent changes in gov purchases Y = C + I + G a permanent increase in government purchases, G reduces consumption, C, roughly one-to-one the change in government purchases and household consumption completely offset each other the following variables do not change: real GDP, Y ; gross investment, I ; the quantity of capital services, κk; the real rental price, R P ; the real interest rate, r; and the real wage rate, w P

17 Useful public services I. How would the model s predictions change if public purchases gave utility to the households? household and government budget constraint would stay the same marginal productivity and depreciation rates would not change conclusion about all variables would stay the same, i.e. government purchases would crowd out private consumption one-to-one, while everything else would remain the same however, households would be happier with this spending as they would derive utility from G for example one unit of G is equivalent in utility to λ 0 units of C λ determines the efficiency of government purchases

18 Useful public services II. λ determines the optimal size of government - this depends on the efficacy of government spending compared to private spending Milton Friedman often asserted that No one spends another person s money as carefully as he spends his own. Barro was once challenged by Friedman to name one government program that provided net benefits in excess of its opportunity costs. Barro replied, national defense, to which Friedman responded, name another. key feature is that national defense is non-rival and non-excludable

19 Temporary changes in government purchases I. Assume that the year 1 gov purchases, G 1 increases by 1 unit, while in all other years G t returns to the original level. The government budget constraint can be described by: G t + V t = T t G t = T t V t = (V t T t ) in the first year the household s real disposable income declines by one unit in later years it does not change our previous analysis tells us that the propensity to consume out of a temporary change in income is greater than zero but much less than one the household will compensate for this income loss by reducing consumption in all future periods by a little

20 Temporary changes in government purchases II. the equilibrium capital services, κk, and real rental rate, R P, does not change: the demand is driven by the MPK, which is unchanged the supply is driven by the depreciation rate, δ(κ), which does not change the real interest rate r = R P δ(κ) does not change the equilibrium labour input, L, and real wage rate, w P, does not change as the labour supply is fixed the labour demand depends on the MPL, which does not change This implies that Y = A F (κk, L) does not change either.

21 Temporary changes in government purchases III. The real GDP does not change: Y = C + I + G real government purchases, G are higher in year 1 by one unit consumption, C, is lower, but by much less than one unit consequently gross investment, I, must fall the decrease in C is relatively small, the decline in I is large year 1 s extra G comes mainly at the expense of I, rather than C when the change in G was permanent, we predicted that most or all of the extra G came at the expense of C

22 Testing the model s predictions by studying the response of the economy to the temporary changes in government purchases that have accompanied US wars. Real GDP far from unchanged

23 Wartime effects on the economy I. Total employment seems to go up during wartime, both military and civilian employment increased relax the assumption that labour supply is fixed large expansion of G negative income effect reduction in consumption and leisure, increase in labour supply magnitude depends on the size of the income effect another problem: C should decrease, but this does not show up in the data L s increases due to patriotism, for the same wage rate and real income people are willing to supply more labour military draft removes men women postpone marriage and/or having children and work instead

24 Wartime effects on the economy II. The labour supply curve shifts right real wages drop. In the data: lower wage rates are supported during WW1, but it is not supported for WW2 and the Korean War. However, in the latter two cases price controls and rationing were in place P understated w P overstated. An increase in the labour input increases the MPK. the demand for capital services curve shifts right as well the real rental price, R P, and the equilibrium capital services, κk, increase In the data: capacity utilisation rate is higher in war times, however the real interest rate is extremely low, even negative. This is a not well understood phenomenon among economists - might be related to the demand for relatively safe assets.

25 Labour supply increases

26 Capital services demand increases

27 Incentive effects of taxes until now: we assumed that taxes and transfers are lump-sum i.e. independent of the household s income and other characteristics in reality governments levy a variety of taxes and pay out a variety of transfers, but none of these are lump-sum now we will allow for taxes to depend on household characteristics, which most of the time are a result of household choices therefore taxes and transfers that are not lump-sum will change the optimal behaviour of households example 1: unemployment insurance - motivates people to be unemployed or to search less example 2: income taxes - discourage people from working taxes and transfers create substitution effects that influence the labour supply, consumption, production, investment, etc

28 Government revenue in the US

29 Federal government revenue in the US

30 State and government revenue in the US

31 Types of taxes taxes on different forms of income individual income taxes corporate profit taxes contributions for social security taxes based on expenditures sales taxes excise taxes customs duties

32 Properties of taxes Average tax the total tax paid relative to the total value Marginal tax the additional tax paid on an additional dollar of income Progressive tax a tax is progressive if the tax rate increases as the taxable base amount increases Regressive tax a tax is regressive if the tax rate decreases as the taxable base amount increases In most countries both the average tax rate and the marginal tax rate is increasing with income.

33 Income tax scheme in the US

34 Percentage of total income tax paid by top tax payers in the US

35 Average marginal income tax rate in the US

36 Taxes in the model - the labour income tax The household budget constraint C + 1 P B + K = w P L + r ( B P + K ) + V T here T contains all the taxes - labour income tax, capital income tax, consumption tax let us look at one particular tax, the labour income tax, and let T w denote the remaining taxes let τ w be the marginal tax rate on labour income this is a big simplification, as in reality there is a whole range of marginal tax rates can think about it as the average marginal tax rate after-tax real wage rate is (1 τ w ) w P

37 The effects of labour income taxes I. The hh budget constraint can be written as: C + 1 P B + K = (1 τ w ) w P L + r ( B P + K ) + V T w by working 1 more unit of time the household can consume (1 τ w ) w P more units hhs can substitute (1 τ w ) w P units of C for one unit of leisure time if τ w rises, then the trade-off between leisure and consumption becomes worse the household will take more leisure time and therefore consume and work less whether there is an income effect depends on what happens w to G = (V T ) = V τ w P L T w if this is unchanged (hence some transfer increased or other tax decreased), then there aren t any income effects if this increased, hence G increased then there is a negative income effect

38 The effects of labour income taxes II.

39 The effects of labour income taxes III. for a given pre-tax real wage rate, w P, a higher τ w implies a lower after-tax real wage rate, (1 τ w ) w P a rise in τ w shifts the labour supply curve leftward from the blue one labeled L s to the green one labeled L s this decrease in labour supply reflects the substitution effect from the higher labour-income tax rate, τ w lowers the equilibrium quantity of labor input, L the marginal product of capital services, MPK, falls, the capital demand curve shifts left (κk) d lowers the equilibrium quantity of capital services, κk real GDP, Y = A F (κk, L) falls as well

40 The effects of labour income taxes IV.

41 Financing a permanent increase in G with a permanent increase in τ w Reminder: when G was financed by lump-sum taxes our finding was that an increase in G by one unit left real GDP, Y, unchanged and reduced consumption, C, by about one unit gross investment, I, labour input, L, and capital services κk were unchanged the real wage rate, w P, the real rental price, R P, and the real interest rate, r, remained the same Now: G is financed by labour income taxes substitution effect: people will work less and enjoy more leisure time negative income effect: as G increases, total taxes increase, income decreases, hence people want to work more net effect depends on the relative strength of these two forces empirically the net effect is very small

42 The Laffer curve T = τ w }{{} w P L }{{} tax rate tax base

43 Taxes in the model - tax on asset income The household budget constraint C + 1 P B + K = w P L + r ( B P + K ) + V T here T contains all the taxes - labour income tax, capital income tax, consumption tax let us look at one particular tax, the tax on asset income, and let T r denote the remaining taxes asset income: r ( B P + K) = ( R P δ(κ)) ( B P + K) let τ r be the marginal tax rate on asset income after-tax real interest rate is (1 τ r ) r = (1 τ r ) ( R P δ(κ))

44 The effects of asset income taxes remember that the interest rate, r, determines the intertemporal substitution for households if the hh reduces C 1 by 1 unit, it can increase C 2 by 1 + r units now there is a tax on asset income, so the trade-off is 1 less unit of C 1, and 1 + (1 τ r )r more units of C 2 if τ r increases, the trade-off becomes worse, so households will consume more and save less in period 1 if we hold G 1 constant, then a higher τ r does not have income effects a higher τ r does not affect labour supply and demand, nor the demand and supply of capital services Y 1 unchanged with a higher τ r C 1 increases, and Y 1 = C 1 + I 1 + G 1 is unchanged, hence I 1 falls one-for-one in the long-run a higher τ r leads to lower I, hence lower K and lower Y