Monday, April 27, 2009

Changes in the funding of the U.S. current account deficit

Since the mid 80's (and with the exception of a small pause in the early 90's) the U.S. has run a current account deficit, which means that domestic spending has been larger than domestic income/production. This deficit was growing in the period leading to the current crisis and it led to concerns about global imbalances and how countries would adjust to them. Here are a couple of charts that show some of this evolution as well as some recent changes that point to adjustments in the way the current account deficit is being financed.

The first chart shows the evolution of the U.S. current account balance. We clearly see the downward trend during the 1995-2007 period and then a reversal during the last year, 2008. This reversal is very much driven by a large drop in imports. Exports have also decreased (as world trade has collapsed), but imports have fallen by a much larger amount. Some of this fall is related to the decrease in the price of oil during 2008 relative to 2007.

U.S. Current Account Balance (Billion USD)

A second interesting fact in the chart above is the behavior of net investment income. Net investment income remains positive during all the years. This means that the US receives more investment income from investments abroad than what it pays to foreigners for the capital that it has borrowed from them. If you take into account the fact that the US has become a large debtor to the world (i.e. that the foreign liabilities are substantially higher than the foreign assets) this is a surprise. One would expect a debtor to be paying interest on the debt. What we see below is that in net terms the U.S. is receiving interest payments on a negative net asset position. In other words, the U.S. benefits from lending at very low rates while it earns substantially higher rates on the capital it sends abroad. Financing a current account deficit under this conditions is much easier!

This investment income has more than compensated the income sent abroad in the form of "Net Transfers". This income include worker's remittances to their countries of origin. The fact that these two variables have been very close to each other means that the current accounts is very close to net exports (the balance on goods and services) [A reminder on balance of payments accounting: Current Account = Net exports on goods and Services + Net Investment Income + Net Transfers].

How was the current account financed during these years? The U.S. was borrowing from other countries (those with current account surpluses looking for investment opportunities). Of course, we observe capital flows in both directions and what matters is the difference between the two. Interestingly, during these years, capital flows in both directions grew. The chart below shows these flows. The current account deficit needs to be financed by foreign lending to the US (labelled below as "changes in foreign assets in U.S." which by convention are positive if there is a flow) in excess of US lending to other countries (labelled below as "changes in US assets abroad" which by convention are negative if there is a flow). 

Funding of U.S. Current Account (Billion USD)
Up to 2007 we see both flows increasing but the size of foreign lending to the US is always larger than the flow in the opposite direction, and this difference funds the current account deficit. 

In 2008 we see a collapse of both flows. The flow of lending to the U.S. goes from around 2 trillion to 600 billion. This collapse is matched by a decrease of capital flows from the U.S. to foreign countries from 1.3 trillion to almost zero. What is even more interesting is that if we split this flow into private and official (government and central bank related) flows, we see that private flows from the U.S. to other countries changed from an outflow of about 1.3 billion in 2007 to an inflow of 480 billion - this represents a change of close to 1.8 trillion. In other words, a large part of the current account deficit in 2008 was financed by U.S. nationals selling their assets abroad and repatriating the funds to the U.S. The change in these private flows more than compensate the drop in capital flows from other countries. [At the same time, official flows from the U.S. to other countries increased to reach almost 500 billion. Most of this lending is likely to be associated to the lending facilities that the Federal Reserve has made available to European central banks]

A final comment on the chart: the "statistical discrepancy" also helped funding the U.S. current account deficit in 2008. The swing from negative to positive from 2007 to 2008 indicates that while in 2007 there were some "missing" capital outflows, in 2008 we are missing some of the capital that flew into the U.S. by an amount that is large (about 130 billion). 

It will be very interesting to see how these numbers change during 2009 and 2010. The selling of U.S. assets is not a sustainable source of funding. It is likely that the current account deficit will become smaller but not by much (given the limited growth that we are seeing in other countries) so there will be a need for capital inflows to the U.S. to be larger than what we have seen during 2008.

By the way, if you are interested in this topic, I recommend the excellent blog of Brad Setser.

Antonio Fatás

Tuesday, April 21, 2009

Estimates of banks' writedowns get (much) bigger

The IMF released today their semiannual Global Financial Stability Report. In the report they estimate the potential loses of financial institutions (writedowns) to be close to $4.1 trillion. Out of those, $2.7 correspond to assets originated in the US, the rest to assets originated in Japan or Europe (of course, the assets could be anywhere in the world). 

Who will incur these loses is a much more difficult question to answer because you need to know who is holding these assets. Nouriel Roubini has estimated that US financial institutions are exposed to about $1.8 trillion losses out of these figures (this calculation takes into account the fact that banks have already written down about $1 trillion of the estimated $4 trillion). How large is this number? Well, if you consider that the total capital of US banks is about $1.4 trillion, it implies that without additional capital, US banks are insolvent.

All these estimates rely on any assumptions about the evolution of the economy, valuation of some assets, so there is a high degree of uncertainty and there is room for things to get better (but also worse...). No doubt that the coming release of results of the stress tests that have been conducted in US financial institutions might bring some interesting surprises and a debate on how to interpret the tests and how to act on them. 

Antonio Fatás

Monday, April 20, 2009

Fiscal policy in action

Here is an interesting chart from the March 2009 interim report of the OECD World Economic Outlook. It calculates the size of the fiscal deficits that OECD countries will accumulate over the 2008-10 period. It decomposes those changes into variations due to automatic stabilizers and those due to discretionary actions (automatic stabilizers are those components of the budget balance that automatically react to the business cycle, such as unemployment benefits).
 
The chart shows that the discretionary actions are the largest in the US (the blue/grey column) measured as % of GDP. However, the overall fiscal stimulus is larger in other countries because of the presence of large automatic stabilizers. The size of the automatic stabilizers is related to the size of the government and the welfare state. 

In the case of Ireland and Iceland, the governments have taken discretionary actions to reduce the size of the deficit, which is the result of a deep recession and the automatic stabilizers (i.e. the government has either increased some taxes or reduced government spending). This "contractionary" fiscal policy is not what their economies need but it is the only way to avoid a loss of confidence.

Antonio Fatás

Wednesday, April 15, 2009

Uncertainty as defined by Larry Summers (or just another example of a two-handed economist?)

In an interview last week, Larry Summers said that "there are risks of both deflation and inflation", so anything is possible... In defense of Larry Summers' lack of commitment when it comes to forecasting inflation, there is indeed an outgoing debate between those who are afraid that advanced economies are heading into deflation and those who believe that the aggressiveness of monetary policy (and the expansion in the central banks' balance sheets) will soon produce a large and persistent increase in inflation rates. 

To some this might sound like the standard two-handed approach of economists but I think that in this case the debate is also reflecting an unusual amount of uncertainty as we are dealing with an episode that is different from what we have seen before (of course, as an economist, I have to defend the profession...). 

There are two reasons for why uncertainty can be so high. First, this is the deepest recession for many advanced economies after WWII. While there are some historical episodes of even deeper downturns (such as the Great Depression), they took place in a very different economic, political and institutional environment. In addition, some of the monetary policy actions that central banks around the world are taking can be seen as an experiment (out of desperation). Traditional monetary policy actions are not enough so they need to try unorthodox measures. Yes, we have a sense on the consequences that these actions will have but we do not have previous historical experiences to produce a good quantitative estimate. If you want to look at how this uncertainty is reflected in differences in inflation forecasts, here is an interesting entry from the macroblog at the Altlanta Fed on the current state of inflation expectations and uncertainty.

The deflation scenario is one that is supported by the data. Today, the U.S. Bureau of Economic Analysis released the CPI estimate for March and on an annual basis, inflation was -0.4%, the first time this measure is negative since 1955. This combined with a weak economy and an unresolved financial crisis could in theory lead to a deflationary spiral. However, one has to be careful reading too much into the figure that was released today. If one looks at the details, the only category where we had deflation was in "transportation", which is driven by the decrease in oil prices over the last 12 months. Falling oil prices is, if any, good news for the US economy. Other items display inflation rates that were clearly above zero on an annual basis (Food and Beverages at 4.3%, Education at 3.2% and Other Goods and Services at 5.7%). The trend looks somehow more worrisome as on a month to month basis we see negative inflation rates across more categories in some of the most recent months, but it is fair to conclude that we are still far from the type of persistent deflation that can cause major damage to the economy.

The inflation scenario is driven by the massive increase in the balance sheets of many central banks (including the Federal Reserve that has probably been the most aggressive of all). We know that this increase in the balance sheets has not translated into a one-to-one increase in the money supply (see earlier entry about the money multiplier), but there is still the risk of inflation if central banks do not react faster and withdraw the additional liquidity when the economy recovers. Simon Johnson has recently written an article in the Washington Post summarizing the inflationary risks of the current monetary policy "experiment". Theoretically, the Fed has the ability to do what it needs to be done to avoid inflation (see, for example, this article by Woodward and Hall on the potential use of the interest rate paid on reserves). 

So what will it be, inflation or deflation? If central banks are good at what they do and they quickly react to the different potential scenarios ahead, we should not see any of the two (i.e. we should see an inflation rate that remains positive but around or below what is considered to be the -explicit or implicit- medium target for central banks). 

Under which conditions can central banks fail? On the deflation scenario, the failure would be one of anticipation. As long as central bank are ahead of the curve and provide the necessary liquidity using whatever means are necessary (including the "helicopter money drop"), they should be able to keep their economies from falling into persistent deflation. On the other side (the inflation scenario), there are potentially two risks: a technical one and a political one. From a technical point of view, central banks also need to anticipate changes in inflation expectations and move as fast as they can to avoid any sustained increase in those expectations. The second risk is more of a political nature. One can envision scenarios where the central bank could be under enormous pressure to give up and let inflation increase above its normal level. If the economy recovery is slow but still fast enough to get inflation expectations increasing, undoing the monetary expansion could have a large impact on the interest rate. An increase in the interest rate will impose a serious cost to governments who are currently accumulating debt at a very fast pace because of large deficits. And if growth is not strong enough to bring the necessary tax revenues, governments will feel the need of either raising taxes or cutting spending, none of which are easy from a political point of view. No doubt that this will be an interesting test of the independence of central banks. We have already seen in recent months central banks deciding on actions that "they did not want to take" but it was necessary because of the "special circumstances". If inflation expectations come back too early, the trade off between inflation and growth (and interest rates) will be very strong. Would it also be considered a special circumstance that requires unusual actions by central banks? Let's hope we do not get there and we do not need to test the independence of central bankers.

Antonio Fatás

Tuesday, April 14, 2009

The money mutiplier has stabilized

One of the most dramatic indicators of the financial crisis has been the collapse of the "money multiplier". The money multiplier is measured as the ratio of the money supply to the monetary base. The money supply is the sum of currency in circulation and bank deposits. The monetary base is the sum of currency in circulation and the reserves held by commercial banks (at the central bank). The money multiplier tends to be above 1 (and that's why it is called a "multiplier") and it is a measure of how much the money supply increases when the central bank raises the monetary base (think about the central bank printing money or increasing the amount of reserves available for commercial banks and what we are after is the final impact it will have on liquidity).

The money multiplier can collapse for two reasons. First, it can be that individuals mistrust banks and they decide to take their deposits away from their bank and keep their liquidity as cash ("under the mattress"). Second, it can be that commercial banks are worried about the future and they accumulate an unusually large amount of liquidity in the form of reserves without lending them to individuals or businesses. This is what we have observed since the summer of 2008, a very large increase in the amount of commercial bank reserves at the central bank and very little lending. The chart below (for the US economy) shows the collapse of the money multiplier at that time. We also see that in recent months the money multiplier has stopped declining although it remains at a very low level (below 1).
 


While the money multiplier was falling, the Federal Reserve increased dramatically the monetary base. This increase translated into a much smaller increase in the money supply because of the falling multiplier. As the multiplier stabilize and very likely starts growing towards its natural level (when banks stop holding large amount of reserves), the money supply will start growing. It will then be the signal to the central bank that it is time to reduce the monetary base in order to keep inflation under control. 

Antonio Fatás

Sunday, April 12, 2009

The difficulty of forecasting around turning points

If forecasting in normal times can already be difficult, doing so during times when the phase of the cycle is changing becomes a real challenge. The amount of uncertainty increases as one can consider scenarios that are very far apart. When in 2007 we saw growth rates decreasing we needed to make a call on whether a recession was coming and if it was, how deep and large the recession was going to be. 

This uncertainty has been reflected in the forecasts that we have seen about the state of the economy in the last 18 months. 

What is interesting about these forecasts is not only that they are hard to make (and that they might turn to be completely wrong) but, in addition, they keep getting more and more pessimistic. In other words, we are only willing to lower our forecasts when we see bad news and the news keep being worse than what we expected so the forecasts are being revised downwards. So it is not just a matter of uncertainty, which could show up as alternating large positive and negative errors. We can think of this process of revising forecasts downwards as a bias in the way we produce our forecasts as the errors are always in the same direction: we are anchored by the past and always too optimistic about the sate of the economy [Strictly speaking, this might not be bias, it could truly be that we receive a string of bad news about the state of the economy that cannot be forecasted, but it is also likely that the consistency of the errors is a sign of our inability to accept that the news are as bad as they look].

Below is a chart with the forecasts for GDP growth rates for 2009 for the World and the US economy as produced by different vintages of the World Economic Outlook (by IMF). Starting with the forecast produced in the fourth quarter of 2007 and until the most recent forecast, we see that the forecast has gone down every single quarter (except for the first). 

It is also interesting that forecast errors tend to be highly correlated across different forecasters. The picture below is from an early study about the accuracy of IMF World Economic Outlook (WEO) forecasts (same source as the chart above). The picture shows that there is a strong and positive correlation between the forecast errors of the WEO and the Consensus Forecasts errors - another common source of macroeconomic forecasts.

And here is a third similar picture with the forecasts of the unemployment rate for the US. Starting in the fourth quarter of 2007 I have plotted the forecast done by the Survey of Professional Forecasters of the unemployment rate 6 quarters ahead. We can see the line shifting upwards signaling a continuous update of expectations, always in the direction of increased pessimism. [The data come from the Federal Reserve Bank of Philadelphia]

In summary, our forecasts keep getting worse before they get better! And whenever we find the new turning point –the bottom of this recession- it is likely that we see a similar pattern but in the opposite direction. Forecasts might remain too pessimistic for a while and are only revised upwards as a continuous set of good news improves our views on the economy. 

Antonio Fatás

Monday, April 6, 2009

Bubbles and Systemic Risk

There is a wide-spread concern that the current monetary easing plants the seeds for the next bubble and therefore for the next financial meltdown. The logic is clear – between 2001 and 2004 Greenspan lowered interest rates to clean up the dot.com bust and as a result of the low interest rates a bubble in the housing market materialized. Over the past few months, Bernanke went well beyond Greenspan’s expansionary policy by lowering rates to 0%, by starting credit “defrosting“ programs as well as by implementing the so-called “quantitative easing”. The ultimate side effect of these efforts to resuscitate the economy must be again a bubble.

Whether a bubble will develop or not in the immediate future is not entirely clear. In this entry, however, I want to use these concerns in order to put up two topics for discussion: (1) Not all bubbles are the same; (2) The big issue is systemic risk – with or without bubbles.

1. Not all bubbles are born equal. The fact that a bubble may develop does not necessarily mean that we will get to the same catastrophic dynamics as in the past two years. Suppose that a bubble develops in the market for gold and the price goes to $3000 per ounce. At some point investors may realize that demand is faltering, they will start selling their holdings of gold and the bubble will deflate. Although for some investors the collapse of this bubble will generate uncomfortable redistribution of wealth, it is not going to lead necessarily to a financial meltdown. Even the dot.com bubble did not generate the same kind of dynamics as the ones we saw after 2007. The gold bubble and the dot.com bubble are very different from the house price bubble. In my view, the key distinction is whether the bursting of the bubble generates systemic risk for the banking sector. One of the fundamental reasons for the severity of the current recession is that credit markets froze because banks had too much uncertainty about their survival probability and they stopped lending. In plain words, the collapse of the bubble had implications for the entire economic and financial system. Not all bubbles have such effects.

Inevitably, there will be more bubbles in the future. There is little doubt that euphoria in some markets at some point will lead to a disconnect between fundamental values and current prices. Whether policy makers should address a bubble and by what means depends to a large degree on the risk of having an economy-wide meltdown once the bubble starts deflating.

2. From regulation to systemic risk policy? The dynamics of the vicious circle between the contraction of the real economy and the disintegration of the financial sector is reminiscent of the dynamics that led to the severity of the Great Depression. Roosevelt arrested these dynamics by implementing four policy changes: (1) financial restructuring (after the banking holiday); (2) expansionary monetary policy (allowed by the suspension of the gold standard); (3) fiscal expansion (i.e. the New Deal; I know that it had a small contribution to the recovery, but at least fiscal policy was not in the way of other policies); (4) regulatory changes (the Glass-Steagall act, etc.). Since the start of the toxic interaction between the financial collapse and the real economy contraction after the collapse of Lehman brothers, I have argued that policy makers have to act in the same four dimensions in order to stabilize the economic activity. This is hardly an original thought, but it is useful to remember that these were also the actions of the Roosevelt government.

At an event organized by INSEAD on April 3, 2009, I made the same argument and in a panel discussion after my presentation, Sir Andrew Large – a former Deputy Governor of the Bank of England – suggested that there is a fifth item that should be added to this list: policy related to systemic risk.

So far systemic risk has been in the realm of regulation. We can think of various restrictions on lending, leverage, capital adequacy ratios, etc., which are designed to minimize the risk of a system-wide collapse. The point of Sir Andrew Large was that these are static measures, but given the sharp increase in financial innovation and the rapid changes in financial markets, shouldn’t we think of a policy body that continuously reviews and changes some of these ratios on a dynamic basis? In the way that monetary policy changes interest rates, the new policy authority will have a set of instruments to deal with developments that raise systemic risk in the economy. Naturally, this body will be in charge of addressing bubbles that have systemic implications. However, it will not be just a bubble-buster, since systemic risk can arise even in periods without bubbles.

There are certainly pros and cons in implementing such a dramatic institutional change and I am sure that I do not do justice here to the ideas of Sir Andrew Large. But I think that this is a very original idea and it deserves further consideration.

Ilian Mihov

Thursday, April 2, 2009

The Euro area is (was) officially in a recession (in case you did not know)

Today, the CEPR (Center for Economic and Policy Research) has announced that the Euro area entered a recession in the first quarter of 2008 (and they identify January as the month when the recession started). This follows the earlier announcement (December 1, 2008) by the NBER (National Bureau of Economic Research) that the US economy had entered a recession in December 2007. This means that both economic areas entered a recession almost around the same time.

In case of the Euro area, this recession signals the end of an expansion that has lasted 57 quarters (or more than 14 years). The previous recession was in 1993. The CEPR does not recognize the 2002/2003 period as a recession although they admit that it was  period of particularly slow growth. In the case of the US, the last expansion lasted 73 months given that there was a recession in 2001 (March to November). 

Both the CEPR and the NBER follow similar methodologies to define a recession, they look for periods of "significant decline of activity" and "activity" is measured by a set of economic indicators (not just GDP). This definition differs from what many refer to as a technical definition of recession, two consecutive quarters of negative growth. As an example, in 2001, the US did not have two consecutive quarters of negative growth as it alternated from positive to negative but the NBER recognizes this period as a recession. 

The reason why their announcement take place months after a recession has started is that they wait for very strong evidence of the change in the phase of the cycle before they make their call. They are not in the business of forecasting business cycles.

More information on the CEPR dating can be found here. The NBER methodology and announcements are on their web site.

Antonio Fatás