[Previous] [Main Index] [Next]

Sunday, September 21, 2003

Some Technical Points in the Last Couple of Articles Clarified: Doubts about 1) Crowding Out Theories, 2) Alleged Low US Savings, and 3) Alleged Panic Sell-off by Foreigners of Their US Financial Assets

Those who've read the previous two articles on US trade deficits and whether they're something to worry and fret about might have encountered some theoretical materials that are worth clarifying here. In particular, what follows tries to throw light on two theoretical claims: 1) Government fiscal deficits crowd out either private investment or US exports (or both), resulting in rising interest rates in the first case or rising US current account deficits in the second. 2) The US economy reflects a problem of low national savings compared to other industrial countries.



NATIONAL SAVINGS: A COMMON CONFUSION

At the outset of the last article, it was noted that there's a frequent confusion made --- you'll find it repeatedly, for instance, in the forum discussions at Brad DeLong's site --- between two different things: the level of US national savings on a net basis (hence minus the cost of replacement capital deducted from gross investment), on one side, and on the other the fact that our national savings aren't sufficient to cover our national investment each year. The latter is true. By definition, the gap is equivalent to the amount of net foreign investment that flows yearly into the US economy on capital account in the balance of payments; in turn, by definition, the net inflow of foreign investment here is equivalent to the size of the current account deficit in the trade of goods and services (plus unilateral transfers, such things as foreign aid or Mexican workers in the US sending money back home to Mexico).


But wait

To say that the US doesn't save enough each year to cover US investment (at home and abroad) does not mean that Americans are necessarily low savers. As the sidebar comment at the start of this article noted, contrary to this common assertion (which rests on some confusion with the gap between savings and investment), Americans turn out in recent international comparisons to be unusually high savers . . . especially on expanded definitions, which include the use of gross investment, a wider view of what public investment amounts to, and the purchases by households of durable goods for repeated use in the future. The latter, just to illustrate it, would mean such things as household's buying cars to drive to work on roads built by government investment.

 

GOVERNMENT DEFICITS AND NATIONAL SAVINGS AND CROWDING OUT

(i.) When calculating national savings, government budgets --- state and federal together --- influence them. If these budgets are in deficit, the deficit subtracts from total US savings. If, oppositely, they're in surplus --- the case of the last 1990s --- then they add to national savings, and hence reduce the need to import foreign capital . . . and in turn, by extension, reduce the size of the current account deficit of the balance of payments. Hence a projected federal deficit for the current fiscal year --- which ends soon --- of $400 billion represents -$400 billion in national net savings. And with state budgets largely in deficit too the fiscal year, by about $90 billion, the decline in national savings here would be on the order of $490 billion. By contrast, a surplus of $100 billion on federal and state budgets would add $100 billion to net national savings and investment --- the two over the year, remember, equaling one another by definition.

If an inflow of net foreign capital to cover a gap between national investments and national savings wouldn't materialize, what would happen?

Interest rates, it's said, would rise in loanable funds markets; investment would therefore fall; and GDP would fall by a calculable multiplier rate in order to bring national savings and national investment into line with one another. In theory, that's all true. In reality, as the previous buggy article on the big trade deficits of the early 1980s showed --- followed, after 1985 remember, by rapidly falling current account deficits and hence, by definition, an equivalent falling level of net foreign investment inflows into the US economy --- interest rates did not rise in the US economy. The 10-year Treasury bond rate, you'll recall, actually continued its steady fall that began in 1982 into the 1990s. It started to fall noticeably, observe, after the sharp, short recession of 1981, which choked out inflationary expectations. And it fell throughout two different periods: rising trade deficits owing to the high dollar of 1982-1985, and declining trade deficits and eventually a small trade surplus afterwards, 1986-1992, when the exchange rate of the dollar was following brusquely. Not only that. In both periods, observe further, the Reagan and Bush Sr federal deficits continued to mount.

(ii.) The theoretical implications here?

Contrary to standard crowding out theories --- which find governmental deficits financed by borrowing from the public (through the sale of new Treasury securities) reducing national savings for private investment and hence raising interest rates that crowd out such private investment; or alternatively, which find rising interest rates here attracting net inflows of foreign capital that raise the exchange rate of the dollar and crowd out US exports abroad (signified by growing trade deficits) --- there is no clear link between either.

What might explain these data's findings at odds with standard macroeconomic theory? Well, consider:

1) Interest rates rise and fall, apparently, according to the public's expectations about inflation on one hand and investment opportunities for savings on the other . . . which lead the public to find ways to come up with more capital to offer in loanable funds markets (to purchase financial assets like equity and corporate bonds and derivatives in financial mortgage markets like Freddy-Mae assets): on that basis, the levels of national savings are more flexible than standard macroeconomic theory presumes. Since the net worth of Americans --- including pension investments and rising equity on houses and condos that we own --- is enormous, on the order of $30 trillion, that means Americans are willing when they want to find ways to increase the levels of savings offered to firms and households looking for capital to buy new equipment, software, plants and offices, and houses and other durables.

2) As for the alleged crowding out of US exports when government budgets turn to deficit --- with rising government deficits closely correlated with rising US current account deficits --- there is no correlation whatever in the data of the last 21 years. The trade deficit fell sharply in the late 1980s and early 1990s when the federal deficits continued themselves to rise. Oppositely, when the federal (and state) budgets turned to surplus after 1997, the US trade deficit continued to leap ahead even more sharply.

 

ARE AMERICANS LOW SAVERS? WIDER VIEWS OF US SAVINGS AND INVESTMENT, COMPARATIVELY STUDIED

(i.) What's the logic behind recent studies like those of Kirova and Lipsey, cited with a link at the very outset of the previous article by Milka Kirova and Robert Lipsey . . . which takes issue with the claim that Americans are low savers and refute it in a variety of ways, starting with a preference for looking at gross investment instead of net investment for calculating the level of investment in the US economy each year and hence, by definition, calculating the level of national savings?

Quite simply, in an era of rapidly changing technologies, the purchase of new equipment and software to replace older hard- and software isn't just a matter of replacing equivalent things. Newer technologies --- embodied in machines and software --- reflect advances and are likely once their users adapt to them effectively to raise the output and thus productivity of the firms and agencies using them. That makes sense, no? By contrast, the accounting technique for arriving at net investment in the new equipment and software is to view these as "replacement" capital, not new capital, and hence to subtract the depreciated value of the older hard- and software from the new purchases. The result? Especially at a time of rapidly falling prices in ICT --- information and communication technologies --- subtracting the depreciated value of the replaced hard- and software from new ones will therefore, it's argued by Kirova and Lipsey and others, markedly understate total national investment on a net basis (minus the value of depreciated capital, remember, that is included in gross investment) . . . just as it tends to reduce the value of gross investment in ICT and software compared to previous years, when the prices were much higher.

The study by Kirova and Lipsey, like others, also broadens the definitions of investment to include a lot of what's considered national consumption --- for instance, rising expenditures on investments in national education or national security (the defense budget) --- and include household durables that have a life of more than one year. It's a good, easily read article full of comparative data, and you will profit from reading it.

 

FURTHER TECHNICAL PROBLEMS OF ESTIMATING NATIONAL SAVINGS IN THE HOUSEHOLD SECTOR (as opposed to the business and government sectors)

(i.) When the claim is made that the US economy is a poor saver internationally viewed, the reference is usually not to the business world but rather to the household sector . . . and, of course, when government deficits are higher, that sector too. The trouble is, the Bureau of Economic Analysis of the Commerce Department --- which handles our national income accounts and estimates household savings --- has to rely on a lot of contestable assumptions and accounting methodologies that don't easily make sense. Consider this problem, a big one: what to do with depreciation of the national housing stock.

What the BEA does is to subtract estimated depreciation of owner-occupied housing from the estimated rental income that owner-occupiers of their houses or apartments receive in principle, with the depreciation and other costs exceeding the estimated revenue. In that case, net savings are adjusted downward for the household sector, and by a large margin. Does this make sense? The truth is that home owners, unlike the owners of apartment buildings who naturally subtract depreciation and other costs from their landlord income, reasonably expect that if they maintain their houses decently, the houses will naturally appreciate in value, not decline; yet as Fred L. Block, The Vampire State (New Press, 1996) --- Block, a well-known left-wing sociologist is using the title ironically --- notes on p. 132, the costs related to home ownership in the US are a large part of the household sector, and the result of the BEA's measures of depreciation "significantly distort reported levels of household savings."

Nor, for that matter, is that the only distortion of the household sector's net savings rate. As noted before, the personal savings of households that go into pension funds and life insurance reserves and other financial assets are usually underestimated by the BEA compared with the institutionalized savings that are reported directly to the Federal Reserve, and again by a large margin.

(ii.) One last observation, very brief, about business-sector investment.

Quite apart from whether it makes sense in an era of rapid technological flux to look at gross or net business investment --- given that new equipment and software embody more advanced technologies that shouldn't be regarded as replacement capital when a firm or agency buys them to replace older equipment and software --- there is another hitch with seeing US business investment as traditionally low compared to, say, Japan or Germany. As the best study on capital productivity in these three countries found --- Capital Productivity, put out by a team of experts at the McKinsey Global Institute in June 1996 --- there are significant differences in the efficiency (productivity) of capital investments in the US as compared to these two competitor economies. Specifically, the productivity was about a third lower in Germany and Japan than here. What does this mean in practical terms? Quite simply, American business can invest at a level one third lower than their German and Japanese firms and be as competitive or more; in turn, that leaves much more of GDP available for American consumption.

All these considerations have to be factored in, then, when you see statements that Americans are low savers, internationally viewed. That claim is wrong.

 



So that previous claim about low savings is wrong. What is true --- a statement with which the claim is confused --- is that our savings are not high enough to close the gap between the level of national savings each year and the intended level of national investment. To close that gap, we rely on foreign capital inflows.

Note that foreigners are happy to invest here --- both long-term (multinational investment) and short-term (portfolio, in American Treasury securities and corporate equity and bonds) --- because of confidence in the American economy's overall soundness and dynamism . . . also, because in the case of Central Banks and Finance Ministries abroad, they want export-led growth for their economies and will buy, if need be, tens of billions of dollars each year to keep their currencies from appreciating against the dollar. That's notoriously the case, as we've shown, with the Japanese and Chinese, but also most of the Asian countries. The upshot? The dollar will rise in value as foreigners invest more here than Americans invest abroad. In particular, it will rise and stay high enough until foreigners are able to sell an excess of exports on current account --- on goods and services and profits expatriated on previous years' investments in the US and elsewhere --- until the current account deficit for the US equals the capital account surplus caused by the larger inflow of foreign capital.

All this, in fact, is true by definition in national income accounts. What is added in the above equation is a causal explanation, widely used in macroeconomics, that explains how the inflow and outflow of capital --- whether short- or long-term, and multinational or liquid --- drives the US balance of payments, with the current account a reactive force what with the huge amounts of capital moving in and out of the US economy every day.

Will There Be Panic Selling Off?

Note one other point. Foreign investors --- private or central banks --- will adjust their portfolios of investments from time to time. On some occasions, they may decide that they can get a better return on, say, investments in Japan or Britain or in the eurozone, and hence they may sell off some of their US investments to that end and exchange the dollars for pound-sterling or Yen or euros to that end. In that case, the dollar will fall against their currencies --- the case of the exchange rate between the dollar and the euro in 2002 and early 2003 (after the opposite for the previous three years had driven the euro down from $1.18 to $0.86) --- unless, of course, as with the Japanese government, its finance ministry will try to keep the Yen from appreciating against the dollar by buying, if need be, up to $300 million daily for months or years. On the whole though, foreign investment has been pouring into the US economy at record levels --- with these ups and downs --- for over two decades; and despite Casandras who vocally fear that some huge financial sell-off will soon occur and force the dollar downward to some sort of dangerous or even catastrophic levels, these fears have proved groundless in the past . . . and will likely to prove so again.

The evidence?

The dollar has appreciated again against the euro the last few months, around about 6% . . . roughly to $1 equaling 1.12 euro. That means the euro, almost 5 years after its began its life at an exchange rate of $1.18, is still 6.0% lower than it started out. And since the IMF projects that eurozone stagnation --- hardly any GDP growth on average the last two years, with next year likely to be even more dismal --- will continue even as the US economy bounds ahead at a very fast clip the rest of this year and next, where exactly would private investors and central banks with dollar reserves invest their money for the best payoff . . . whether short-term or long? In Japan? In China? In the German or French or Italian stock markets? Some sure. But to foresee some unsustainable level of US current account deficits as a trigger force for a major sell-off of their US financial holdings ---- which would, among other things, undermine Asian export-led growth into the huge booming US economy as their currencies rose in value against the dollar, while further hindering eurozone exports to us as the euro rose rapidly in value too --- seems far-fetched, with no solid grounding in the experience of the previous two decades.

More specifically, the fears that the dollar will fall in value soon --- say, on the order of 25% - 50% --- and bring widespread harm, such as rising interest rates here, slower growth, and inflationary pressures as the dollar price of imports rises --- clashes with the experience of 1986 - 1994, when the dollar fell by 50% or more against the Yen and German Mark (to which almost all the EU countries' currencies were tied). There was no noticeable harm to the US economy: inflation rates didn't rise in this period, they came down steadily; interest rates didn't rise, the 10 year long-term Treasury bond rate fell steadily as the dollar fell in value and hence net foreign investment inflows fell too; and though GDP growth fell, it was largely for cyclical reasons after 7 years of fast growth (including 1986 and 1987), not because of rising interest rates as foreign capital fled the US economy: no such rise occurred. Meanwhile, by 1991 and 1992, the US current account deficit had turned to a surplus. All the while, US federal deficits were rising (until 1992), with no crowding out noticeable in loanable funds markets or in US exports. These latter doubled in value after 1985.

 

Were Other Countries Hurt by the Rapidly Falling Dollar?

Most economists, it seems safe to say, would have predicted lots of harm, just as lot like Brad DeLong continue to worry about it happening in the future if the dollar should fall sharply again . . . what with the reliance of Japan and Germany and of much of other EU countries too on export-led growth. The reality after 1985, a 9 - 10 year period of a rapidly falling dollar?

No, or little, harm at all. Just the contrary.

Specifically, if you look at GDP statistics, you find that after 1985 --- precisely as the dollar declined and the US current account improved noticeably --- both Japanese and German GDP rates leapt ahead, and markedly so. They did so until 1991 or so, at which point the Japanese ballooning stock- and real-estate markets crashed, a collapse from which the Japanese economy has never recovered . . . and the German Central Bank had to put on severe brakes on money growth to choke out new surging inflationary pressures that had resulted in the previous year from given 18 million East Germans hard West German Marks for worthless East German Marks at an exchange rate of 1:1! The outcome? Cash-rich, the East Germans went on a frenzied buying spree, and overall unified German GDP was soaring at an unsustainable rate, resulting in those inflationary pressures. Since then too, it's worth noting, the German economy hasn't recovered either. It has emerged as the worst grower in the EU, and it bears many of the signs of an over-statist, over-rigid economy that Japan has struggled with for a decade, with belated reform efforts under way in both.

Why weren't export-led countries like Germany and Japan hurt after 1986 by the fast fall of the dollar against their currencies?

That's a tough question. Here we can only conjecture. By 1986, presumably, the revival of their economies fueled after their stagnation in the early 1980s by an export drive into the booming US economy after 1982 was generated more domestic-led consumption and investment as driving factors of GDP growth. On top of that, all of Asia was now booming, and its economies were big sources of new exports from Japan and West Europe (and the US). By 1990, too, Latin American countries were overcoming a decade of stagnation too, and most were booming as well . . . a further source for exports from Europe, Japan, and the US. There may be other explanations. Only a detailed study of the sources of GDP growth in Germany, the rest of the EU, and Japan could clarify this, including what was happening to their export sales abroad to various foreign markets.