Economics 456. International Macroeconomics and Finance: Section 4. Geoffrey Dunbar. UBC, Winter February 15, 2013

Transcription

1 Economics 456 International Macroeconomics and Finance: Section 4 Geoffrey Dunbar UBC, Winter 2013 February 15, 2013 Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

2 Balance of Payments International borrowing and lending has three possible advantages for a country. Consumption smoothing Efficient investment Diversification of risk. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

3 Balance of Payments There are often practical reasons for international borrowing and lending. Climate shocks such as hurricanes or natural disasters such as earthquakes typically require sizeable re-investment. International borrowing can reduce the costs of such investment. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

4 Balance of Payments As another example, consider a developing country that wishes to invest in it s education system. If the country is still poor, raising money via taxes is costly in welfare terms since the marginal utility of consumption is high. Borrowing and repaying in a future period is welfare improving if the returns to education are greater than the borrowing costs. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

5 Balance of Payments One big question is how much can a country borrow? In this section we will derive the long-run budget constraint. A key restriction we will impose is that there is a no-ponzi scheme. We will also consider the effects of shocks, risk and productivity. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

6 Balance of Payments Before we begin writing down a model, we should define exactly what we are trying to measure. A countries balance of payments comprises two accounts: the current account and the financial account. The current account measures a country s net payments and receipts for imports and exports (trade balance), net payments and receipts of income (net factor income from abroad) and net unilateral transfers. The capital account measures the net change in a country s financial assets that transact as gifts. The financial account measures the net change in asset exports minus asset imports. (Not everyone distinguishes between these two, they used to be called just the capital account.) Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

7 Balance of Payments The wonderful part of the balance of payments is that: current account = financial account capital account To see how this works, consider the purchase of a Samsung Galaxy Phone by someone in Vancouver for $600. The phone is imported from South Korea so imports rise by $600. However, since the payment is in dollars then Samsung is receiving a $600 financial asset (the currency), thus the financial account falls by this amount. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

8 Balance of Payments The textbook covers some basic macro aggregates. GNE Gross National Expenditure: is equal to GDP plus the trade balance. GNI Gross National Income: is equal to GDP plus net factor income from abroad. NUT Net unilateral transfers: net transfers received (given) from (to) the rest of the world. GNDI Gross National Disposable Income: GNI plus NUT. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

9 Balance of Payments The distinction between GDP, GNI and GNDI can lead to different conclusions regarding an country s economic progress. For example, Ireland paid a sizeable amount of it s income to foreigners via factor payments. Factors can be labour or capital. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

10 Gross National Income Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

11 Net Unilateral Transfers Net unilateral transfers are likewise relatively unimportant for developed countries but they matter for developing countries. One of the main types of transfers is official development assistance or other charitable gifts, e.g. the Bill and Melinda Gates Foundation. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

12 Gross National Disposable Income Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

13 The US Current Account Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

14 Current Account and Savings Although the current account relates to the national income identity: Y = C + I + G + CA, it is important to remember that the current account is also the difference between savings and investment: CA = S I Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

15 Global Imbalances To frame the issue of global imbalances, recall that savings is simply foregone consumption. It can, in theory, be used for consumption at any moment and thus there can be large swings in savings. Also note that savings can be either private saving or government savings: S = S p + S G Perhaps it is not surprising that government savings can be volatile. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

16 Global Imbalances Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

17 Global Imbalances Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

18 Global Imbalances The issue with government deficits can be framed through the lens of Ricardian Equivalence. For government deficits to matter, it must be that Ricardian Equivalence does not hold, otherwise private savings would adjust to offset government deficits. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

19 Global Imbalances Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

20 The Financial Account We can think of the financial account as: net export of domestic assets + net import of foreign assets We have seen that the financial account, the capital account and the current account are related. Although the balance of payments accounting implies that these sum to zero, this is not often true in practice. Part of the problem is accounting for financial derivatives. These are often included as part of the statistical discrepancy. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

21 Balance of Payments Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

22 External Wealth What we have examined so far are income flows. However, flows over time become stocks. It should be no surprise that countries can amass wealth in foreign assets. A country s external wealth is the difference between it s holdings of assets in the rest of the world and the rest of the world s holding of its assets. Note that there can be a currency mismatch. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

23 External Wealth Changes in external wealth come from the financial account and valuation effects, VE. The valuation effects are due to capital gains or losses which can be due to interest rate changes or exchange rate changes. If we call the change in external wealth W then: W = FA + VE = CA + KA + VE So countries gain external wealth through thrift, gifts or valuation changes. External Wealth is also known as the Net International Investment Position (NIIP). Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

24 External Wealth Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

25 Valuation Effects Can Matter! Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

26 The Long-Run Budget Constraint The long-run budget constraint (LRBC) of a country is a limit on how much debt it can borrow. In this examination of borrowing constraints we are only interested in feasibility. This means we are only interested in whether a country can repay. This does not imply that it will choose to repay. We have already used examples of LRBC when we examined exchange rate models. Here we shall focus on these constraints in more detail. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

27 The Long-Run Budget Constraint: Assumptions Prices are perfectly flexible. The country is a small open economy The interest rate on all debt is r, the world real interest rate. There are no NUTs. The capital account is always zero. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

28 The Long-Run Budget Constraint Because of our assumptions, the change in wealth each period must come from the current account. Here it is the trade balance plus net factor payments from abroad. Thus, W t = TB t + rw t 1 where t is the year and TB refers to the trade balance. Note that rw t 1 are the factor payments to the external wealth. Note as well that because prices are perfectly flexible and we are using a real model then there are no valuation effects. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

29 The Long-Run Budget Constraint Since W t = W t W t 1 then rearranging the last expression yields: W t = TB t + (1 + r)w t 1 Now note that since we are deriving this equation without reference to a specific t then we can also write: W t 1 = TB t 1 + (1 + r)w t 2 This is our usual repeated substitution trick. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

30 The Long-Run Budget Constraint Now recall that what we are trying to find is the maximum amount a country can borrow and still repay at some point in the future. Let s rewrite the expression above as: W t 1 = TB t 1 + r W t 1 + r And now use repeated substitution to replace W t. We will get: W t 1 = TB t 1 + r + TB t+1 (1 + r) 2 + TB t+2 (1 + r) W (1 + r) Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

31 The Long-Run Budget Constraint This is the same as the LRBC in the textbook, Equation 6-1. To see this, just multiply both sides of the above expression by 1 + r. (1 + r)w t 1 = TB t + TB t+1 (1 + r) + TB t+2 (1 + r) W (1 + r) This equation is a bit more interesting than perhaps the textbook suggests. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

32 The Long-Run Budget Constraint W (1+r) Note that 0 in either of two ways: W = 0 or (1 + r) faster than W. Also notice that it is the assumption of a constant r allows us to write the constraint this way. Also note that this budget constraint is hard to forecast: is a particular TB too negative or will future TB s offset it? Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

33 The Long-Run Budget Constraint The LRBC also has implications for national income and national expenditure. Recall that: GNE + TB = GDP. So we can substitute in for the trade balance. (1+r)W t 1 = GDP t GNE t + GDP t+1 GNE t+1 + GDP t+2 GNE t+2 (1 + r) (1 + r) Our observations from the last slide matter here too. As a pre-cursor to some of the analysis we will do, consider the following situation. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

34 The Long-Run Budget Constraint Suppose that GDP is produced used capital and labour effort. And suppose that, for whatever reason, a country is allowed to borrowing a very large amount in a period so that GDP-GNE is negative. Now consider the perspective of the borrowing country. To repay the debt, the country must either save more or work more. These are costly to utility. Or the country can default. This is presumably costly too since the country s future borrowing will be affected. So the decision to repay or not involves a trade-off between future disutility and current disutility. To throw even more into the mix, suppose that it is an OLG environment. Then the current generation which is deciding to repay or not may not be alive to experience the disutility in the future. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

35 The Long-Run Budget Constraint A related but somewhat different situation is also interesting to consider. Suppose that you are the leader of the government and you are concerned that future gov ts may make spending decisions of which you do not approve. One thing you could do to tie the hands of a future government is to leave office with a large amount of external debt. This will effectively force GNE GDP and so no future government s will have the fiscal room to enact new spending. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

36 The United States and Exorbitant Privilege Recall our discussion of the simplifying assumptions. One thing we assumed is that the interest rate was constant for both assets and liabilities. This is not always true in practice. For the US, the interest rate it pays on it s borrowing is historically percentage points lower than it earns on it s foreign assets. (Perhaps this is explained by investor s perceptions of a flight-to-quality.) Thus, even though the US is a net debtor it actually earns net factor income from abroad. This is unlikely to continue for several reasons, most particularly the fact that it may be increasingly difficult to find new buyers for US debt and also because development in the rest-of-the-world should lower the marginal returns to such assets. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

37 The United States and Manna fomr Heaven A second difference for the US is that it has routinely enjoyed capital gains on its external wealth. These capital gains are effectively transfers from the rest of the world to the US. The average return from 1980 to today is roughly 2 percentage points a year. This combined with the interest rate differential gives the US roughly percentage points higher return on its external wealth than other countries. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

38 Balance of Payments Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

39 Balance of Payments Let s reconsider our LRBC that we derived (somewhat) theoretically a few slides ago. Do the valuation effects that we observed for the US have any implication for the LRBC arguments? Yes! The valuation effects imply that the external wealth position is not simply due to the trade balance and net factor payments. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

40 Balance of Payments Of course, the US may be a special case. So perhaps we might want to consider more evidence before we become skeptical of the LRBC concept as a whole. Let s consider emerging markets. The LRBC assumes that all countries have access to borrowing and lending at the world interest rate. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

41 Balance of Payments Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

42 Balance of Payments Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

43 Balance of Payments These graphs suggest that emerging markets face different interest rate schedules and face different access to capital markets then advanced countries. Emerging markets must pay a different risk premium than advanced economies. This suggests that the interest rate schedule is not constant for these countries and depends, at least in part, on the path of trade balances. Second, emerging markets are more likely to face a sudden stop in which it s financial account shrinks rapidly. This implies that the current account balance must similarly shrink. So how can one model a sudden stop in the LRBC? Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

44 Balance of Payments So now that we have examined the LRBC we must ask the question: why? What role does the LRBC play for a country? The LRBC is a constraint on consumption. Thus, to investigate the effects of the LRBC for an economy, we need to describe the consumption path for an economy. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

45 Balance of Payments In general, we shall assume that countries exhibit consumption-smoothing preferences. Before we begin on this track, I also note that it is not entirely clear that the LRBC is unaffected by consumption patterns. There is what we might call an endogeneity problem. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

46 Balance of Payments Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

47 Balance of Payments So, if it is the case that countries choose the level of financial openess because of their preferences over consumption smoothing then it is also difficult to know how to model the LRBC as a constraint. Basically, in this case the shape of the constraint depends on the desired consumption path and then we would need to model this. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

48 Consumption Smoothing and the Current Account The textbook example (on page 216) considers the case where borrowing is offset by consumption changes in all future periods. In a model where agents have preferences for consumption smoothing and must repay this is optimal. Suppose the current account in a particular year falls by Q. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

49 Consumption Smoothing and the Current Account To see the role that the LRBC plays for consumption, let s examine how a change in income matters for the current account. Recall, (1+r)W t 1 = GDP t GNE t + GDP t+1 GNE t+1 + GDP t+2 GNE t+2 (1 + r) (1 + r) Now suppose that, for whatever reason, GDP falls by an amount Q in period t but is expected to return to its original level in the subsequent and all future periods. Now, consider the case where agents prefer smooth consumption. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

50 Consumption Smoothing and the Current Account Clearly, the agent will prefer to borrow internationally to offset the fall in domestic income. Otherwise consumption would necessarily fall and this isn t optimal given his preferences. Ideally, the agent wants to reduce consumption in all periods by an amount C. Thus it must be that C < Q. This is the implication of consumption smoothing. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

51 Consumption Smoothing and the Current Account So for consumption to not fall in period t, it must be that the country borrows Q C. This allows the agent to only reduce consumption in period t by C. This fall in external wealth must be offset by future surplusses so that the LRBC still holds. Because the agent wishes to smooth consumption, the future trade surplusses must reduce consumption in all future periods by C. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

52 Consumption Smoothing and the Current Account Using the LRBC, it must be that: Rearranging: 0 = ( Q C) + C 1 + r + C (1 + r) 2 + C (1 + r) Q C = C 1 + r (1 + 1 (1 + r) + 1 (1 + r) ) Using our infinite series representation: Q C = C 1 + r ( 1 1 1/((1 + r) ) = C 1 + r (1 + r r ) = C r Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53

53 Consumption Smoothing and the Current Account So we can now solve for an expression in terms of consumption. We can solve this to yield: r( Q C) = C C = r 1 + r Q This is the fall in consumption in response to a fall of Q in GDP assuming consumption smoothing preferences and that the LRBC is satisfied. Geoffrey Dunbar (UBC, Winter 2013) Economics 456 February 15, / 53