The Open Economy: A Complex System of Many Moving Parts

For the United States, our economic framework is that of a large, open economy that must include the potential influences of economic and policy changes in foreign countries.² Capital mobility and a floating exchange rate are characteristics of our U.S. financial framework. Owing to these two market characteristics and generally market-determined interest rates, we judge our central bank as an independent institution. The central bank is independent not only in the political sense, but also in the economic sense given its ability to set policy without significant institutional constraints such as the legal one-target inflation restrictions as imposed on several central banks such as New Zealand.³

Under such a monetary/financial regime, an easing of monetary policy would typically lead to a decline in interest rates and an expansion in economic growth and incomes. Ceteris paribus, we would also suspect, under a flexible exchange rate regime, a weaker dollar and capital outflows from the United States. But all other factors are not constant. This is what makes our economy so dynamic and what excites us in the challenge of economic analysis.

While monetary policy is one factor that influences asset prices, there is also fiscal policy. In recent years, we have witnessed U.S. fiscal policy turn restrictive in terms of the pace of spending growth, which would evoke a decline in aggregate demand, prices, and interest rates, as well as a weaker dollar. That expectation turned out to be only partially correct, as we only saw interest rates decline, while aggregate demand, prices, and the dollar exchange rate have continued to rise. Finally, beyond the patterns of economic policy, there is the underlying flow of economic activity that reflects the factors of labor force growth and productivity gains.

Outside the Frictionless Economic Model

Yet these stylized results are within the context of a perfectly smooth economic system. In reality, we do not have that perfectly functioning system. Instead, our work in this book has consistently emphasized the existence of partial adjustment mechanisms, imperfect information, and the operation of an economy with no real set of “equilibrium” conditions.

Since 2008, easier monetary policy has been the primary means to achieve faster growth in the U.S. economy. This policy has produced lower interest rates and helped provide for the basis of improved economic output. Lower interest rates reduced the burden of financing real estate and raised the valuation of both bond and equity assets. Yet the gains in housing have been less than many anticipated given the decline in interest rates. In part, this has been due to the greater regulatory constraint on lending institutions, which reflects the many factors not accounted for in a perfectly functioning, purely private-sector model. Meanwhile, financial markets, especially the equity market, are said to have climbed a wall of worry. Why?

International capital flows and their impact on asset prices are central to understanding the global framework that we accept would more accurately represent the actual conduct of economic activity and helps explain U.S. equity and fixed-income market behavior.

We focus on several points. First, economic activity reflects the impact of the change in relative forces/prices that mimics Newton’s third law—and the choices households and business make in response to changes in those forces. Second, the economy is in constant motion and not at an equilibrium point. There is no Aristotelian natural state here. As a result, even when the domestic U.S. economy may give an appearance of balance, foreign shocks, such as the 2011–2012 euro sovereign debt crisis or the collapse of global oil prices since mid-2014, lead to changes in the domestic economic and financial activity. There is a duality of internal-external economic conditions that moves markets.

Another central theme to the analysis is that economic activity does not fully adjust to a new equilibrium—instead, economic activity reflects a series of partial adjustments given the constraints of imperfect information and the frictions associated with labor and capital. Moreover, as we have previously noted, different economic actors, such as workers and employers, exhibit differential speeds of adjustment for prices and exchange rates and this gives rise to a cyclical pattern in profits.⁴ Finally, the role of exchange rate expectations remains critical, especially in the world of financial assets—equity and bond valuations reflect the expected future, not past, returns for these assets.

Looking at the World We Have

Let’s return again to our fundamentals of disequilibrium and dynamic and partial adjustment under conditions of imperfect information, and the importance of expectations in the economy. The entire period of 2005 to the present (2016) has been characterized by a sense of disequilibrium in the economy. As the economy was characterized by a housing bubble prior to 2007, a real expansion was followed by an economic bust, a long period of persistent high unemployment, slow recovery in housing starts, and subpar aggregate income growth. Contrary to the perfectly competitive conditions of economic theory, households, and businesses do not have perfect foresight on the effectiveness of monetary/fiscal policy or the underlying pace of the real economy. As a result, households and businesses have made only a partial adjustment to the change in expectations.

In a repeat of the experience of 2002–2004, the initial monetary policy easing post-2008 did not lead to an immediate economic recovery. This led many analysts to question the traditional effectiveness of monetary policy. In addition, fiscal policy was initially extremely expansionary with the 2009 stimulus program. However, a significant share of that program was not well directed and much of the program was spread out over time. Again, the impact of economic policy actions was less than anticipated. Finally, there was no perfect information about the potential change in the underlying trend growth of the economy as well as the impact of increased regulatory structures on the financial system. In recent years, the concern about labor force participation and its impact on potential growth has moved to center stage.

Dynamic adjustment is exemplified by the gradual and sometimes stop-and-go movements of households and business decision makers. While the economy improved on many fronts, individuals have been hesitant, as evidenced by the gradual improvement in consumer confidence (Figure 10.1). For businesses, confidence surveys indicate that business sentiment is not back to prerecession levels (Figure 10.2). Both the National Federation of Independent Business (NFIB) and Wells Fargo indices denote a long period of below-average business confidence and only a slow recovery back toward prior heights. Finally, policy uncertainty provides a basis for understanding the lack of follow-through effects from fiscal and monetary policy since 2008.

Money and credit expansion by the Federal Reserve did lead to a decline in interest rates and an expansion in income, but these patterns did not immediately lead to the expected real results in the economy. Imbalances persisted, including excess unemployment in the labor market, strong capital inflows in the currency market (despite the low interest rates) and a rising dollar (again despite a lower interest rate regime for short-term securities). In a frictionless model, when interest rates decline we should witness currency outflows and a weaker currency. However, even in a frictionless model we witness capital outflows and exchange rate depreciation if other economies cut interest rates more on a relative basis. Exchange rate depreciation would reduce the relative price of domestic goods, raise the demand for domestic output, and thus national income would expand. This would increase the demand for financial assets (money, equity, and bond assets) and thereby alter their respective values. The relative improvement in the balance of trade via exchange rate depreciation should raise domestic output and employment.

However, in the United States’ case, easier monetary policy was accompanied by lower short-term interest rates as expected, but what was unanticipated were the greater capital inflows and rising currency values. The capital inflows led to rising asset valuations for equities, bonds, and real estate (Figure 10.3).

But in economics, unbalanced forces bring further change—and the seven years following the recession have witnessed a continuous set of unbalanced changes both internally and externally among countries. Whatever depreciation we have witnessed for the euro and Japanese yen has not been accompanied by an economic recovery in each region and has not led to an outperformance of financial assets in each region relative to the United States (Figure 10.4). Initially during the current economic expansion, easier monetary policy in the United States led to lower interest rates, dollar appreciation, stronger capital inflows, and rising asset values for equities, bonds, and real estate. Yet even with the switch to tighter U.S. monetary policy relative to continued easier policy abroad, we have witnessed lower U.S. long-term rate and rising asset values for both bonds and equities. Positive financial market reactions to both easier and tighter U.S. monetary policy signals the complexity of markets and their reading of economic fundamentals.

The Rising Visibility of Foreign Economic Policy

Foreign monetary policy has become an increasingly important channel driving U.S. asset prices. Changes in expectations of foreign monetary policy have become a factor on their own, independent of an actual change in policy. Changes in policy expectations act as shocks much the same as any exogenous economic shock such as an oil price spike.

For example, note the recent volatility in expectations of the European Central Bank, where weak responses to easier policy led to increased expectations for further policy easing.⁷ In another recent article, foreign investors were portrayed as piling into U.S. Treasury debt due to the “broad demand for safe government debt amid global turmoil and uneven economic growth.”⁸

The list of economic shocks includes changing expectations of fiscal policy in the Eurozone, alterations of economic goals in China, and the on again/off again Japan consumption tax along with easing by the Bank of Japan. As evidenced by this litany, foreign economic policy, as well as U.S. domestic economic policy, contains a significant degree of policy uncertainty and these actions, or lack of action, affect not only the pace of economic activity but also the volatility of activity.

Imperfect Capital Mobility—A Constraint to Equilibrium and an Invitation to Volatility

Beyond the actions of public and private policy makers, market structure is also a source of frictions in the economy. Financial regulation and the consequent market structure from such regulation limits the flow of capital into countries and investment into the various possible set of real and financial assets. These frictions reinforce the message that the realities of the marketplace contrast with the models of perfectly smooth-functioning capital markets that drive many predictions of future asset values.

To begin, many central banks take actions to alter the market determination of exchange rates. Second, countries differ by their level of sophistication in their capital markets. For example, sovereign debt is a significant portion of private bank assets in Europe and this led to further complications when such sovereign debt was downgraded and asset values deteriorated. This limited the extent of economic recovery in Europe, reinforcing the message that the impact of monetary and fiscal policy actions is limited by the character of the marketplace in which such actions occur. Capital flows are clearly limited in cases, such as China, where the currency is not convertible. In cases such as the United States and Europe, increasing capital requirements for private banks have the result of redirecting funds away from funding private activity to funding public debts. Along these lines, there remains the legacy of legal risk associated with prior action associated with the 2008–2009 Great Recession. In the marketplace, the existence of all these regulatory frictions results in a misallocation of credit away from the allocation that would be determined in the private marketplace and gives rise to dead weight losses in the economy.

Asset price movements reflect the framework of an exceedingly complex, imperfect marketplace. Capital flow frictions, the uncertainty of future policy actions, and differential speeds of adjustment make long-term economic projections, within the context of short-term horizons of the next election, enormously difficult. These limited forecast horizons and heavily discounted future returns would tend to discourage longer-term investment projects. As a result, asset prices would be anticipated to be lower and more volatile in such a context. Moreover, asset prices would not tend to equalize across countries. Interest rate differentials would persist and a less efficient allocation of capital across countries would continue over time.

Three Challenges

Our perfect market model faces three challenges. First, information is imperfect, not only for the economic data but particularly for policy maker intentions. A persistent challenge in recent years is that the pace of inflation has actually turned out to be much below policy maker and market expectations, while the impact of policy actions has also delivered real economic growth below expectations.

Second, exchange rate and interest rate adjustments are not instantaneous and often drag out over time. Moreover, dynamic adjustments may also be delayed or diminished by financial and economic regulations that inhibit markets from reacting quickly to initial economic shocks. Therefore, there can be a period of persistent disequilibrium in the marketplace. We will see this below in the Plaza/Louvre examples. Third, price adjustments are far from uniform across sectors and often contradictory to our initial expectations as illustrated by recent exchange rate/interest rate actions.

For decision makers, the importance of these issues is illustrated vividly in the figures below. International revenues for the S&P 500 equities (Figure 10.12) are significant, particularly so for sectors such as information technology, materials, and health care. Therefore, equity market valuations will be impacted by all three challenges relative to economic expectations. We saw earlier the volatility of private capital inflows into U.S. financial assets. The rapid expansion of capital flows post-2008 was followed by a slowdown and then another increase during the challenges of the euro crisis of 2012. This pattern of financial flows provides a significant impact on the values of U.S. financial assets and once again the ability of domestic firms to finance economic activity.

The experience of the Plaza and Louvre accords of the 1980s provide a glimpse into the challenge of a view of market-determined exchange rates and dynamic adjustment in the marketplace in the context of discrete policy actions. Domestic U.S. producers were unhappy about the large and ongoing current account deficits for the U.S. in the mid-1980s. However, the policy reaction in Washington was not immediate. At the time, exchange rate policy favored leaving currency markets to set exchange rates and was termed benign neglect.¹⁰ Yet over time, political pressures overcame market philosophy as benign neglect with respect to dollar’s value was dropped in September 1985.

The decision to intervene reflected a change in policy makers’ reaction function, in part due to political pressures. Policy makers then adopted a proactive stance. This outcome reflects the evolution of a public-policy agenda that finds market results unfriendly.

To appreciate the implications of our three challenges to market and policy processes, we now turn to the perfect market model and the workings and results implied by such a model.

A Model for Policy—Not the Real World

Our model structure begins with three assumptions. First, the money stock is the driving force for economic activity and is under the control of the monetary authority. Second, there is a floating exchange rate regime. Third, the balance of payments is indeed balanced.

In the case of a small country (one in which the country does not set the benchmark for global interest rates), a monetary expansion is effective in the sense that an expansionary monetary policy raises output. This effective policy operates through two channels. First, the monetary expansion alters the relative price of domestic goods versus foreign goods. Second, the monetary expansion alters the relative interest rates between countries and, in an integrated world, capital markets lead to changes in global rates, exchange rates, and capital flows.

In our model, output is a function of interest rates and real exchange rates. A real depreciation of a currency raises demand for domestic goods as global demand for a country’s output rises. Meanwhile, a rise in interest rates leads to a fall in demand and output. Within the monetary sector of the economy, real money demand depends on real income and the rate of interest. To close the model, we require that the domestic rate of interest equals the given rate of interest in the world.

How does our idealized model work? A monetary expansion results in a rise in real money balances since prices are sticky in the short run. This channel highlights the importance of both the dynamic adjustment and price adjustment assumptions. With the rise in real balances, the market adjusts by a fall in interest rates and thereby an expansion of income.

At this point, there is disequilibrium as domestic interest rates are now below global rates, which will produce a capital outflow, a balance of payments deficit, and thereby lead to exchange rate depreciation. In turn, exchange rate depreciation lowers the relative price of domestic goods, leading to an increased demand for domestic production; income grows to restore monetary equilibrium at the world interest rate. The trade balance improves due to the relative decline in the price of domestic goods and consequently imports. The result is a monetary expansion that increases output through improvement in the balance of trade.

In a perfectly competitive economy, we can define internal balance as an economy at full employment in the labor market. For external balance, the balance of payments reflects the influence of the flexible exchange rate structure. The current account balance serves as the target in a free market model—a contrast from the persistent surplus/deficit positions of countries in the current global economy. Under that current system, a country aims neither to borrow nor lend excessively from abroad.

In addition, we have an asset market equilibrium that will be altered when interest rates are not equalized across countries. In this case, there is another dynamic adjustment critical to market prices. As a rule, the differential between interest rates between countries will be offset by anticipation of exchange rate appreciation/depreciation between currencies. Investors are considered to be indifferent between similar securities (denominated in domestic/foreign currency) if the differential in interest rates is exactly offset by the expected exchange rate change.

An interesting result here is that interest rates in the home country are no longer fixed as a change in interest rates abroad implies a change in domestic rates as well. Domestic interest rates are low relative to foreign rates if the market expects an exchange rate appreciation of the domestic currency. In contrast, the domestic interest rates are high relative to foreign rates if the exchange rate is expected to depreciate. This result helps explain the persistent low interest rates in Japan/United States, while the yen/dollar appreciates, and the persistently high interest rates in some emerging markets, such as Brazil, as its exchange rate is expected to depreciate.

Expectations of appreciation/depreciation reflect the difference between the perceived long-run equilibrium rate and current actual exchange rate. In the perfectly competitive model, we have three economic conditions at equilibrium. First, the money supply will equal money demand. Second, interest rates will adjust for depreciation/appreciation and be equated internationally. Finally, expectations will be a function of the discrepancy between the long-run and current exchange rate. For example, if the current exchange rate is above the perceived long-run equilibrium exchange rate, then interest rates will decline. In contrast, if the actual exchange rate is below the long-run exchange rate, then interest rates will rise.

There is also a dynamic aspect to the adjustment process. The rate of adjustment of prices in an economy depends on the structural links in the economy as well as how expectations are formed. For example, the actual rate of depreciation of a currency is proportional to the discrepancy between the long-run equilibrium value of the exchange rate and its current value. The speed of adjustment depends on the responsiveness of real and monetary factors to the change in interest rates.

The perfectly competitive global financial market that frames the vision many observers have of global markets begins with three simple assumptions—all of which are violated daily in the practice of markets and policy decisions.

Sharp adjustments in exchange rates, capital flows, and long-term interest rates are often not set in the marketplace by changes in economic fundamentals but by discrete policy changes at the national level. We now turn to the implications of these violations and how to identify them in empirical work.

Violation of Three Conditions for Perfect Competition

First, there is imperfect information on both the policy and underlying economic model. We have already examined the role of imperfect information in economic factors such as growth, inflation, and consumer spending. There is also imperfect information about the intents of policy makers—what is their policy reaction function?

In 2015 and early 2016, central banks in both the euro area and Japan have engaged in surprise policy actions within a very imperfect global marketplace and produced counterintuitive results for capital flows and inflation. Experience in both the euro and yen situations cited earlier are the product of notably different conditions than what the traditional perfectly competitive model would indicate. The Bank of Japan went to negative rates and yet the yen revalued upward. As for the Swiss National Bank, it altered its reaction function by dropping the peg to the euro and adopting negative interest rates.

Imperfect Information and Price/Policy-Maker Discovery

Decision makers face two problems of imperfect information. First, the current indications of nominal interest rates and exchange rates may not reflect market forces and would be distorted by administered interest rates and targeted exchange rates by sovereign authorities. As a result, capital flows between countries are driven by a distorted set of interest rate/exchange rate combinations. Second, the current model for policy making is unclear.

Imperfect information creates problems where market price signals may not reflect the economic tradeoff between alternative investments or exchange rates. Financial markets may also have truly poor information about the policy framework and intentions of policy makers—the market shocks due to the SNB delinking the Swiss franc from the euro, as well as the surprise BOJ move to negative rates in January 2016 and then surprising the markets by not moving at all in April. Even earlier, we contemplate the market disruptions associated with the U.K. withdrawal from the European Exchange Rate Mechanism (ERM) on September 16, 1992.

Persistent Disequilibrium in Prices and Exchange Rates’ Dynamic Adjustment Problem

Things take time—especially in economics. There is a distinct dynamic adjustment following any economic shock or policy initiative. Markets do not adjust instantaneously and therefore, there is a period of disequilibrium as markets move to a new equilibrium.

Portfolio investment in U.S. fixed-income instruments provides a valuable illustration. Purchases of U.S. Treasury, corporate and agency debt continue to fluctuate and give the appearance of constant adjustment toward a desired equilibrium and yet never seem to get there. The volatility of these purchases indicates a constant stream of market/policy surprises and responses. At the global level, the volatility of portfolio investment flows for the United States, United Kingdom, and Canada (Figure 10.14) provides a similar impression of constant motion.

Lags in market responses to policy initiatives are illustrated in Figure 10.15. In this case, the Plaza accord was intended to weaken the dollar versus the Deutschmark and yen. First, the dollar had actually peaked in late 1984/early 1985 prior to the Plaza accord in September 1985. The Louvre accord was intended to stabilize the dollar but that did not occur right away. The dollar continued to decline in value until early 1988. The Plaza/Louvre experience also illustrates the challenge of overshooting, which is discussed later in this chapter.

A free, competitive market model assumes that any imbalance in trade and capital flows would be resolved by changes in exchange and interest rates. But in the real world, policy makers have incentives to avoid adjustments for domestic political reasons.

Price rigidity gives rise to an incomplete adjustment in the rest of the economic system. When exchange rates/interest rates are out of line, then there is either too much or not enough output/employment in various sectors. Market prices for goods and services are also out of line. The mispricing leads to investment in overpriced sectors and not enough investment in underpriced goods and service sectors. Rigid nominal interest rates/nominal exchange rates lead to inflation/deflation that alters the terms of trade and therefore the balance of imports/exports. This can lead to involuntary unemployment of workers and capital in one country and overemployment in another. This sets up a day of reckoning when the adjustment does occur. This is one of the issues faced by China in 2016 as its exchange rate adjusts and the Chinese authorities attempt to guide a transition from export-oriented goods production to more domestic services production.

Dynamic adjustment—adjustment over time to shocks—particularly the wholesale revaluation or devaluation of a currency is often frustrated by domestic policy for domestic political reasons. Domestic sectors that have benefited from a mispriced currency pay a high price over the long run when the currency barrier breaks down. In the short run, domestic firms are protected from competition by an undervalued currency—but there are two problems. When protected by an undervalued currency, there is an overallocation of both business capital and labor to the protected industry. When the inevitable currency adjustment occurs, then the doors are open to large business failures and structural unemployment. There is also the problem of capital controls, which directly misallocate capital among economic winners and losers.

As illustrated in Tables 10.1 and 10.2, we first look at the behavior of capital flows, both direct (Table 10.1) and portfolio (Table 10.2), for various nations. We want to identify patterns of behavior in a series to decide if these patterns return to equilibrium or whether they are in persistent disequilibrium over time. Our econometric results suggest that the investment behavior (both direct and portfolio) is not mean reverting for all countries. That is, if decision makers believe that capital flows move around a long-run average in their respective countries, then they are incorrect. Furthermore, decisions based on the past averages are misleading and emphasize the issue of incorrect information.

Incomplete nominal adjustment of exchange rates and interest rates alters the path of capital flows and trade and frustrates the allocation of resources to best uses across the globe. Undervalued currencies tend to protect domestic firms from global competition and allocate too many resources to those firms and generate overemployment. Alternatively, overvalued currencies fail to protect domestic firms and workers and create involuntary unemployment or overemployment. For many nations, these imbalances lead to the political issues today.

During the early Bretton Woods era, the persistent disequilibrium of exchange rates prevented price discovery for both exchange rate and interest rates. The period was characterized by persistent trade deficits (United States and United Kingdom) and surpluses (France) and a steady claim on gold from deficit countries by surplus countries. These flows eventually were resolved by a break in the gold exchange–based system and sharp adjustments in exchange rates and interest rates.¹³

Price Adjustments: Limits by Policy Makers

Policy sets the context—and the risk. When economic policy is not oriented toward flexible, free markets but rather toward interventionist policies (e.g., administered interest or exchange rates, limited capital flows), policy actions increase the risks of sharp breaks in market prices.

Economic outcomes are not simply a reflection of perfectly competitive market forces but the impact of given policy regimes and their alterations. Such policy actions lead to often sharp market adjustments. Most recently we have seen a sharp policy change by the SNB (Figure 10.16). By delinking the Swiss franc from the euro, the SNB has broken patterns for all three economic activities: delinked the exchange rate peg, entered into negative interest rate territory, and altered the incentives of capital flows into the Swiss franc.

Flexible exchange rates and interest rates—jointly determined in the private marketplace—also reflect policy actions. As illustrated in Figure 10.17, the actions of the central banks of Denmark and Sweden to alter their exchange rates also meant altering their interest rates and venturing into the world of negative interest rates.

Here is a clear illustration of the interrelationship between market prices that creates interdependent price movements for exchange and interest rates and also creates a dynamic interplay of exchange rates and interest rates.

In Figure 10.18, the three-month forward exchange rates are illustrated for four cross rates. High interest rate currencies tend to show a depreciating forward curve. Low-rate currencies tend to show an appreciating forward curve, or at least that is what interest rate parity would imply in a frictionless market. For the 2002–2008 period, the three-month-ahead forward rates for the euro/dollar continued to decline, indicating that euro interest rates would be high relative to U.S. interest rates. After 2008, these expectations changed and U.S. rates were anticipated to rise relative to euro rates.

After the spike in 2008 associated with the global recession, the dollar relative to the yen initially declined but has risen since 2012. Here, then, is an alteration of market expectations. Whereas initially the dollar/yen signaled declining interest rates for the United States, the period since 2012 indicates rising interest rate expectations relative to Japan. A similar pattern occurred for the British pound versus the dollar since early 2014 as well as for the U.S. dollar versus the Korean won.

The interrelationship between market prices and expectations gives rise to the phenomenon of overshooting.¹⁴ When a monetary authority eases policy, under conditions of perfect capital mobility, interest rate differentials are offset by exchange rate movements. Domestic interest rates exceed the global interest rate only if the domestic currency is expected to depreciate in real terms at a rate equal to the interest rate differential. When market expectations are introduced the situation is altered.

Now, the initial exchange rate response to an economic surprise is greater than the long-run response. When the central bank eases, the currency initially depreciates, but given the price level, the currency falls below the long-run value. That is, the real exchange rate has not changed since relative prices do not initially adjust. Therefore, the interest rate differential must now equal the anticipated change in the exchange rate. When the current exchange rate is below its long-run equilibrium, then the exchange rate is projected to appreciate over time.

While overshooting may be a short-run phenomenon, we are also interested in permanent change. In Tables 10.3 and 10.4, we focus in on the possibility of a permanent shift in our interest rate regimes for a number of countries. The results are stark. Interest rates are not mean reverting. This presents two problems. First, a shock to the system often implies a permanent shift, not temporary overshooting, in interest rates away from the initial equilibrium. Second, some shocks lead to a shift but not a permanent change in the interest rate regime.

Capital Flows and Treasury Yields in the Post–Great Recession Era

Our global economy is more open today than it has ever been. While this has many positive implications for growth and efficiency, openness also makes decision making more challenging, as private and public policy actions in one part of the world affect private- and public-policy makers’ decisions in other regions. In addition, the changing nature of the economic ties between countries requires a continuous reevaluation of the benchmark relationships. That is, the degree of economic interdependency of a country (e.g., the United States) may increase with some countries (after the North American Free Trade Agreement [NAFTA] with Canada and Mexico) and decrease with others (e.g., trade sanctions with Russia or Iran).

In this section, we continue the theme of identifying the interactions of capital flows and their influence on asset prices around the globe. Our goal is to help decision makers characterize the differential behaviors among countries and asset classes in the ever-evolving open world economy.¹⁵

We characterize U.S. capital inflows behavior using an annual data set for the 1975–2012 period. The Hodrick-Prescott (H-P) filter–based trends for total inflows and direct investment intimate that the pace of capital inflows has slowed down since the early 2000s. In addition, our analysis indicates that the three measures of the global economy that we focus on here (real GDP, inflation, and the current account balance) exhibit different behavior for the post-2007 period compared to the 2002–2007 era.

Tracking U.S. Capital Flows: What Has Changed?

The H-P filter–based trends for total capital inflows (Figure 10.19) and direct investment (Figure 10.20) have been moving upward since 1975.¹⁶ However, during the past 10 years, both trends have begun to show signs of flattening out, hinting that the pace of capital inflows has been slowing since the early 2000s.

The mean, standard deviation, and stability ratio for each series is reported in Table 10.5. In all cases, the stability ratio is greater than 100. Yet a structural break test indicates no evidence of a break for all series except other private holdings, which had a structural break in 2008 (Table 10.6). However, outliers are present in all series, which is consistent with the higher stability ratios of these volatile series. As a means of testing for structural breaks in these series, we conducted Augmented Dickey-Fuller (ADF) tests on each series. The ADF test results indicate all series are mean-reverting. In sum, statistical results indicate the U.S. capital inflows series are largely mean reverting, but there are some volatile periods when the current account series moves away from the mean.¹⁷

	1975-2012			1975-1989			1990-2012
Variable	Mean	S.D.	Stability Ratio	Mean	S.D	Stability Ratio	Mean	S.D.	Stability Ratio
Official Portfolio Holdings	–47.30	618.73	–1308.13	–237.84	804.80	–338.37	68.68	454.46	661.67
Direct Investment	24.70	57.22	231.63	36.90	53.62	145.30	17.28	59.22	342.71
Total Inflows	22.64	65.50	289.36	25.56	38.83	151.92	20.85	78.23	375.13
Private Portfolio Holdings	–222.93	1097.72	–492.41	28.69	44.89	156.47	–376.08	1380.50	–367.07
Other Private Holdings	28.22	119.15	422.23	55.77	123.81	221.99	11.45	115.74	1011.04
Other Official Holdings	370.25	2064.20	557.52	866.40	3328.24	384.14	68.24	418.07	612.64

Source: U.S. Department of Commerce

Official Portfolio Holdings (Mean Reverting)
Break Date	Type of Break	Coefficient
Jan-86	Additive Outlier	–3010
Jan-90	Additive Outlier	1873
Jan-99	Additive Outlier	–1006
Direct Investment (Mean Reverting)
Break Date	Type of Break	Coefficient
Jan-93	Additive Outlier	160
Jan-84	Additive Outlier	135
Jan-04	Additive Outlier	121
Total Inflows (Mean Reverting)
Break Date	Type of Break	Coefficient
Jan-10	Additive Outlier	306
Private Portfolio Holdings (Mean Reverting)
Break Date	Type of Break	Coefficient
Jan-91	Additive Outlier	–5827
Jan-10	Additive Outlier	–3310
Other Private Holdings (Not Mean Reverting)
Break Date	Type of Break	Coefficient
Jan-08	Level Shift	–103
Jan-97	Additive Outlier	225
Other Official Holdings (Mean Reverting)
Break Date	Type of Break	Coefficient
Jan-86	Additive Outlier	12378
Jan-92	Additive Outlier	1355
Jan-95	Additive Outlier	1244

Are Global Sovereign Yields Mean Reverting?

To characterize the global bond market, we utilize four major countries’ 10-year sovereign yields: the United States, Germany, Italy, and the United Kingdom. The H-P filter–based trend and log of the U.S. 10-year Treasury yield is plotted in Figure 10.21. The H-P trend moved steadily downward from 1994 to 2012, but has rebounded since then. The uptick in the trend is consistent with the ending of quantitative easing (QE) programs by the Federal Reserve Board.

Trends for Germany (Figure 10.22) and Italy (Figure 10.23) have also generally been downward since 1994. In contrast to the United States, both countries’ trends have plunged in the past few years, which may signal that investors foresee that current expansionary monetary policy may continue in the near future and the economic recovery will continue to be disappointing. The downward trends of the German and Italian sovereign yields are consistent with the announcement of QE from the ECB. The trend in the U.K. 10-year gilt yield (Figure 10.24) has a pattern consistent with the U.S. Treasury trend, bottoming out in 2013 and moving upward in 2014. Overall, the H-P filter analysis shows the global Treasury market has a mixed trend, as two major markets (United States and United Kingdom) were trending toward higher rates while the other two (Germany and Italy) were anticipating further stimulus, which kept rates low.

The mean, standard deviation, and stability ratio for each country’s sovereign yields are reported in Table 10.7. For all four countries, the stability ratios are less than 50, which indicates sovereign yields were notably stable during the complete sample period (1994–2014) and subsamples. In the final step, we determine whether measures of global Treasury market are mean reverting (Table 10.8). All four sovereign yields experienced structural breaks and are nonstationary, indicating these series are not mean reverting. Thus, investors should not assume sovereign yields in these nations will return to any sort of long-run average.