Chapter 2

Must There Be a Lost Decade?

A Socratic Dialogue with the President Explains Why Not

The U.S. economy faces two quite different yet interrelated challenges between 2010 and 2050. The first challenge stems from the prospect that the current decade becomes a Lost Decade. Worries in this regard center on the likelihood that subpar growth will make it difficult for the U.S. unemployment rate to fall from its current rate back to a more normal rate. The present chapter shows how a fundamental rethink of fiscal policy and of the role of government can simultaneously raise GDP growth, reduce the unemployment rate, slash the fiscal deficit thus placating bond market vigilantes, and redress the crisis of deteriorating U.S. infrastructure. A new kind of Marshall Plan is proposed for the entire decade and beyond.

The second challenge facing the nation centers around the longer-run fiscal crisis of ballooning entitlements spending on Social Security and Medicare that threaten national solvency between 2020 and 2050. Chapter 3 will propose concrete solutions to this second challenge. In the case of the all-important Medicare spending crisis, the proposed solution is original and, to the best of my knowledge, has not been proposed anywhere else up to now.

The solutions proposed to both the shorter-run and longer-run American crises are deduced from a few basic assumptions that almost anyone will find “reasonable.” This being true, the resulting policy prescriptions should be compelling to citizens and policy makers of many stripes, and thus should prove win-win in nature. As a result, it should be possible to build widespread support for these policies and in doing so to dampen today’s Dialogue of the Deaf about government deficits, and to break policy gridlock.

The U.S. economy is still recovering from its most serious slump since the 1930s. In doing so, it faces a number of unusual headwinds that have given rise to fears of a Lost Decade between 2010 and 2020. Primary concern centers on a level of economic growth inadequate to bring down the nation’s unemployment rate (properly measured) to normal levels. Additionally, substandard growth will also prevent the decrease in the fiscal deficit now being demanded by the global bond market. Finally, there is the growing negative impact on GDP of the nation’s deteriorating infrastructure. Just as positive infrastructure investment raises productivity and hence GDP growth (witness China), so can negative net infrastructure investment blunt productivity growth and thus economic growth. In this regard, both the United States and the United Kingdom are increasingly criticized for their dilapidated infrastructure and lack of remedial investment. Both nations need to put a new roof on their house, and to do so soon.

The principal goal of this chapter is to show how refinement of current macroeconomic policy can help to resolve all these concerns and thus avoid a Lost Decade. More specifically, policies are identified that can simultaneously reduce the fiscal deficit and reduce the unemployment rate and significantly increase economic growth and productivity and placate anxious bond market vigilantes and confront our infrastructure investment needs on a massive scale. In short, Mr. President, we can have our cake and eat it too. This is a view completely at odds with today’s widespread assumption that we must either shrink today’s deficits way down and thus impair economic and employment growth, or else continue with unprecedented large fiscal stimulus and hope against hope that bond market vigilantes will “understand.” It turns out that this either/or assumption is fallacious, and has blinded us to the prospect of a brighter future.

This chapter first reviews the seven principal headwinds to a strong economic recovery to date. This review will orient the material that follows. The discussion then goes deeper. Utilizing two quite different GDP forecasting perspectives, it demonstrates why strong growth is very unlikely to materialize under current policies during the remainder of the decade. Yet very strong growth will be required to bring the unemployment rate back down to normal. The final section of the chapter sets forth the revisions in fiscal policy required to achieve all four goals identified earlier. To make the discussion more accessible to the reader, I have written this last section as a hypothetical Socratic dialogue between President Obama and myself. I have found this Q&A format to be extremely effective in simplifying the story being told, and in permitting the reader to participate by asking questions and getting timely answers as these questions unfold in a natural logical progression.

Seven Headwinds That Have Stunted U.S. Economic Recovery

Before identifying these headwinds, consider Figure 2.1. This shows the magnitude of recessions and recoveries during the past 70 years, where we measure magnitude by the percentage change in GDP.

Figure 2.1 Annualized GDP Change from 1929 to 2010

Note: Data are annual from 1929 to 1946 and quarterly from 1947 to the first quarter of 2011.

Data source: Bureau of Economic Analysis, and Strategic Economic Decisions, Inc.

Figure 2.1 shows the magnitude of the recession from several different perspectives. First, the total drop of output in the 2007–2009 recession was larger than in any previous recession during the past half-century. The magnitude of this drop can best be visualized as the amount of black ink lying below the zero growth line. Simply compare this most recent drop with those of previous recessions. Second, contrast the magnitude of the recovery (the black ink above the zero line from 2009 on) with the loss of output during the recession. The ratio of this recovery/recession magnitude is the smallest in recent times. This is why it is taking so long for GDP to regain its prerecession level, and why the job market recovery has been so tepid. Third, by reviewing the entire time span shown, you can glimpse one of the great macroeconomic developments of the twentieth century, namely the stabilization of life on Main Street. If you use formal statistical analysis, you can show that the decade-by-decade volatility of GDP declined by over 75 percent over the century (earlier decades not shown). There are five interesting reasons why this occurred, and these are reviewed in note 1.1

The magnitude of our most recent recession represents a sharp deviation from this trend toward milder recessions, and the principal reasons for this will be discussed in Chapter 4. The principal culprit turns out to have been excessive leverage by households and banks alike. Paradoxically, the widespread willingness to assume so much leverage resulted in part from the very stability that people became used to after World War II as the great moderation of economic volatility occurred. Life simply was not as risky as it had been, for all the reasons cited in note 1.

So what were those seven headwinds that have prevented a more rapid and indeed normal recovery during the present business cycle?

1. Depressed Labor Market

There are two distinct ways in which the current U.S. labor market is the worst since the Great Depression of the 1930s. First, there is the length of the unemployment line, and second there is the rate at which the line becomes shorter during a recovery. As for the total level of unemployment, the unemployment rate reached a shocking 16 percent by late 2011, two years after the onset of a recovery boosted by the largest fiscal and monetary stimulus in half a century. I am using here the broadest possible unemployment measure known as the U6. This measure includes part-time workers along with discouraged workers who have stopped looking for jobs. The principal reason for the length of this line was the very large number of people fired during a very deep recession, along with the ongoing accumulation of new entrants to the work force (students and immigrants) who failed to find employment since 2007.

Another reason was the high growth in productivity, a blessing in all times other than a recession. During four quarters, productivity growth exceeded 4 percent. The higher productivity growth is, the easier companies find it to lay off workers when demand falls off during a recession. The result will be an even longer unemployment line. A final reason for high unemployment was the magnitude of distress in the housing industry, and also in state and local government.

The rate at which the unemployment line becomes shorter depends primarily upon the strength of the recovery (e.g., the rate of GDP growth), which, as we just noted, has been substandard. It also depends upon the rate of productivity growth. Fortunately, productivity growth stopped rising in 2011. This is now helping to dent the unemployment rate. Even so, the GDP growth rate in the recovery has averaged about 2.5 percent, and new entrants have continued to join the workforce every year. Accordingly, even with much lower productivity growth, companies have had little need to rehire workers. In most past business cycles, the economic growth rate was much higher during the first few years of recovery and the productivity growth rate was somewhat lower than this time around. As a result, the cyclical fall in unemployment during upturns was much more robust than it has been this time.

2. Household Spending Restraint

After over three decades of excess borrowing and binging, households are finally cutting back. Household debt is being paid off, even if mortgage foreclosure and bankruptcy is one reason why. The savings rate as computed by the Commerce Department soared to just under 6 percent from less than 2 percent before the crisis. While Americans remain American, and thus consumer spending has recovered to a certain extent, consumer sentiment is very different than it was before the recession. People have discovered the true costs of being over-leveraged, most especially in housing, but also in general.

The combination of soaring mortgage default rates, personal bankruptcies, and joblessness has been chilling to almost everyone. Additionally, Americans are aging and discovering how bleak their retirement prospects really are. The median net worth of heads of households that are 65 years old is currently one-fourth the level needed for people to retire as they had planned. As a result, millions will attempt to remain in the workforce into their seventies, and will be forced to cut back on their lifestyles.

Soaring medical costs are playing a role as well. Everyone has friends who find themselves crippled with astronomical hospital bills, even if they have insurance. Indeed, with insurance companies requiring ever-higher copayments, underinsurance is becoming as grave a problem as lack of insurance used to be before ObamaCare. Finally, the reduction in household net worth as a whole has sobered many families. An increasing number of people do not expect their net worth to bounce back. The resulting “wealth effect” is yet another headwind for the economy.

3. Housing Industry Depression

Investment in new housing has been a mainstay of the U.S. economy since 1945, but it is contributing little if anything to the current recovery. Indeed, the housing crisis worsened as 2011 unfolded, with house prices once again falling, and the inventory of unsold homes rising in many parts of the country. The number of annual housing starts fell from about 2,068,000 at the start of the bust in 2006 to only 581,000 today—over two years into the recovery. It is likely that residential investment spending will continue to prove a drag on the economy for many years to come. Perhaps the most important reason why lies in the belief of an increasing number of Americans that “our house may not in fact be the best investment we ever made.” This represents the inversion of a basic article of faith in American life. People at long last are learning the frightening downside of leverage.

4. State and Local Government Contraction

The crisis in state and local government is the worst in over half a century, reflecting a collapse in revenues and the advent of the day of reckoning after decades of giveaways to public employees’ unions. Consider the performance of the public sector during the past three recessions. In the 36 months after the start of the July 1990 recession, the net change in state and local employment was up by 678,000 jobs. In the 36 months after the start of the March 2001 recession, the net increase in jobs was 665,000. Thirty-six months after the start of the recent December 2007 recession, the number was negative 148,000. It would rise to over negative 600,000 by October 2011. This is as shocking as the data on housing starts. Until this recession, the public sector was regarded as a safe haven in the employment world.

There are two unusual drivers of what has happened in this sector. First, the long-predicted explosion in medical and retirement costs among public sector employees is occurring. Payment of these benefits has state-constitutional priority in many states over payment for the services (running the subways) that the public needs. So workers who provide these services have been getting laid off in record numbers. Second, tax receipts dropped more than expected. One reason why was the trend toward lower income taxes for poorer citizens and higher taxes on the rich. As it happened, the misfortunes of the rich proved highly correlated when the bust occurred, so that tax revenues collapsed more than anticipated. This further increased the need to shed jobs.

5. Lackluster Business Investment

Even though many companies are flush with cash and have had stellar earnings, their capital spending has not propelled the economy forward, as many hoped it might, and as it did in the 1995–2000 period. To begin with, businesses expected that the fears and dampened spirits of many households would contribute to a sluggish recovery, just as they did. Second, capacity utilization of the nation’s capital stock has been low so there was little need to invest rapidly. Third, investment opportunities in the newly developing nations have been greater than here at home, and corporations have vigorously pursued these instead of investing at home. Fourth and finally, surges in capital spending usually occur just after some important technological breakthroughs have occurred. These tend to occur without warning, and none seem to be in sight at present.

For example, the rewiring of America during the 1995–2000 capital spending boom was largely due to the wave of Internet activity following Marc Andreessen’s launch of the Netscape browser in late 1994. No one was caught off guard more than Microsoft’s Bill Gates who promptly declared this browser to be the technological game-changer of our time, and who proved correct. The ensuing boom in capital spending was a record breaker. Tax receipts soared so much that a U.S. fiscal surplus was achieved, unemployment fell to under 4 percent, and productivity growth soon exploded. Alas, such investment booms are infrequent, and government policy cannot trigger the dates of their arrival.

6. Oil, Commodities, and Health-Care Price Shocks

The rise of gas prices to around $4.00 per gallon by mid-2011 had depressive effects on consumption. While no one can know how the Middle Eastern crises will play out, there is a significant probability of further turbulence—even in Saudi Arabia and Iran—such that oil could ultimately spike to price levels hitherto unknown. This would indeed prove a headwind. For the moment, today’s global slump has driven prices back down.

Middle Eastern instability and supply disruptions are not the only forces impacting the oil market. There is also soaring demand from the developing economies, which is not being matched by enough new supply for prices to remain moderate in the future, except in recessions. The same is true for numerous other commodities.

The growing proportion of total world economic growth attributable to fast-growing developing countries ensures that this supply/demand imbalance will continue for several decades to come. Traditionally, increased supply would have matched or exceeded increased demand, keeping prices stable despite strong growth. The history of commodity prices in the twentieth century attests to this point. What is new is the lack of incentives for companies with the required know-how to invest in those thugocracies where so much of the world’s copper, oil, and other commodities can be found. As a result, increased supplies that traditionally would have been forthcoming will not be. Copper offers a very good case in point.

As for health-care costs, little need be said. In poll after poll, Americans express grave concern not only over soaring health-care premiums, but also over the phenomenon of underinsurance. Who does not know someone who discovers that their insurer will cover only 65 percent of the cost of treatment? With U.S. surgery and hospitalization costs by far the highest in the world, and rising, who can afford a 35 percent copayment? This often will wipe out a lifetime of savings.

7. Inevitable Reduction in Fiscal and Monetary Stimulus

During most recoveries, interests rise and fiscal deficits fall. In principle, both developments provide a drag on recovery. The good news is that such drag is more than offset by a resurgence of pent-up demand and optimism in the household and business sectors. But this time around, the magnitude of the cyclical adjustment back to normalcy in the years ahead is unprecedented. Interest rates start off much lower than ever before, whereas fiscal deficits are humongous. In an era of already restrained growth, the withdrawal of such large fiscal and monetary stimulus could prove to be yet another headwind to the rapid recovery the nation needs to achieve its goals in output and employment. Yet unless great care is taken, the bond market may soon demand precisely such a reduction in stimulus.

Reasons for Lackluster Growth During the Remainder of this Decade

The headwinds we have just discussed explain why GDP growth during the current recovery has been lackluster, far below the level of growth required to reduce the unemployment rate back to normal levels. Some of these developments imply a continuation of subpar growth and hence a truly Lost Decade—for example, the need for fiscal and monetary tightening, and consumer pessimism. Yet to do justice to this risk, we must now delve deeper and look further ahead in an attempt to forecast the period 2013–2020. To do so, let me sketch two different approaches to GDP forecasting. While interrelated, they are quite different in spirit, and permit two quite different routes to a forecast. When applied to the remainder of this decade, both presage a level of growth inadequate for the nation’s goals of significant deficit reduction and employment growth.

GDP Forecasting Proper

Every beginning economics student will recall the textbook definition of GDP as

GDP = Consumption + Investment + Government Spending + Net Exports (2.1)

where Consumption is household spending; Investment includes corporate investment in plant and equipment as well as household investment in housing; Government Spending represents all government spending excluding transfer payments, and Net Exports is the difference between what the nation exports versus imports. The more goods the nation makes and exports to others (compared to what it does not make and imports), the greater GDP will be. Clearly, the greater the value of each of these four terms, then the greater GDP will be.

To prepare a forecast of GDP growth, namely the future percentage change in GDP, most analysts utilize this adding-up equation and assess what the percentage change will be (positive or negative) in each of the four components. By then weighting these four subsidiary forecasts by the relative share of GDP accounted for by each component (Consumption being the largest at about 71 percent), and by adding the four results, they arrive at the desired forecast of GDP growth itself. What can we now forecast along these lines for the remainder of the decade?2

Keeping in mind the various headwinds detailed previously, it is reasonable to expect consumption growth to be restrained given the public’s general shift toward caution, if not outright pessimism. To be sure, consumer sentiment has improved, if unevenly, since the dark days of 2008 and 2009, and Americans are not depressive by nature. Nonetheless, the drag of growing concerns about the need to save more, to deleverage, and to confront soaring medical costs will depress spending and living standards. Business investment growth will also be restrained for a considerable period. This reflects the three developments detailed earlier. The same applies to residential investment. It will certainly improve as time goes on and today’s inventory of unsold homes is gradually drawn down. The deeper reality is that fewer and fewer people believe that buying a house is a good investment, as was already noted. To make matters worse, many cash-strapped baby boomers will be forced to unload their houses sooner rather than later.

The third component in equation (2.1) is Government Spending. Given growing demands of the global bond market for fiscal discipline, given growing fiscal hawkishness by average people who now read weekly about future national bankruptcy, and given the new power of the Republican Party in Washington, the fiscal deficit of 10 percent of GDP will have to shrink, largely by curtailed government spending. But by how much? IMF and World Bank officials insist that fiscal deficits exceeding 3 percent are unsustainable in the long run. To put it differently, deficits exceeding this level will excite the ire of the bond market and result in higher bond yields, which will depress growth, making matters worse. Reducing the deficit from 10 percent of GDP to 3 percent represents a de facto reduction in the growth of GDP itself by 10% – 3% = 7%, other things being equal.

Even though such fiscal contraction would be spread out over time, its cumulative effect would still be depressive. Moreover, the political pain triggered by such fiscal austerity makes it unlikely that politicians will actually make the required cuts, unless somehow forced to do so by currency and/or bond market vigilantes. The United States is not yet in that position, so Washington is now hoping to get away with longer-term deficit reduction of about 3.5 percent of GDP, not 7 percent. There is a heated debate as to whether the nation will even achieve this goal (which would leave the annual deficit at about 10% – 3.5% = 6.5% of GDP), and whether the bond market might find such a deficit too high and ultimately go on strike anyway.

Whatever the case, the magnitude of deficit reduction that will be required one way or the other will be much greater than during any past recovery, and thus the Government Spending term in equation (2.1) will be a drag on GDP growth. Historically, a deficit of 2.5 percent of GDP in good times might have risen to 4.5 percent during a recession. This would have amounted to a stimulus of 2 percent of GDP during recession. The deficit would then fall back by 2 to 2.5 percent of GDP during the recovery as tax revenues rose and unemployment payments fell, representing a modest fiscal contraction.

Finally, there is Net Exports, always the great white hope for rekindled growth. The problem here is that every nation wants to boost domestic growth by boosting net exports. Yet since any one nation’s gain from increased net exports will be offset by another’s corresponding loss from lesser net exports, it is rare that a given country ends up restoring domestic growth from greater exports alone. (Recall here that the sum of all nations’ individual trade balances must be zero.) Given U.S. addiction to imported oil, with higher long-run oil prices looming, it is hard to see a significant improvement in Net Exports, although the low level of the dollar on a trade-weighted basis may help exporters and tepid consumption growth may restrain imports.

A Tentative Forecast

When all these observations are quantified, the resulting GDP forecast for the remainder of the decade lies between 1.5 and 3 percent, with a mean of around 2.25 percent. This assumes, of course, that there is no adverse shock that might cause another recession during this period. On the surface, GDP growth of this magnitude may not seem so bad. Indeed, it is higher than we are likely to witness in Europe or Japan. However, appearances are deceiving. The “recovery growth rate” averaged over all business cycles of the twentieth century was 6.5 percent. This was the growth rate required to rehire idled workers of all kinds in a nation with a fast-growing workforce and a good growth rate of productivity.

With today’s unemployment rate remaining very high, the 2.25 percent growth rate implied by the forecast based on equation (2.1) is woefully inadequate. The result could well be a “new normal” unemployment rate of 14.5 percent measured on a U6 basis, or 8 percent as measured by the more familiar headline statistic. This in turn would be sociologically unacceptable given the very modest safety net for those out of work in the United States, and given that American society is all about work. Indeed, the first question someone asks when meeting a new person in the States is “And what do you do?” Nowhere else is this true.

An Alternative Approach: National Income–Based Forecasting

Scientists and philosophers often speak of “duality” as one of the deepest and most counterintuitive aspects of reality. Thus, for every elementary particle in physics, there must exist its dual, an antiparticle; for every positive number there will be its negative; for a male there is a female; for light matter there is dark matter; for a coin there is a head and a tail; for market-clearing prices there must be market-clearing quantities in microeconomics; for good there is evil; and for fairness there is unfairness in ethics. The basic idea here is that there exists a hidden unity or “organicity” in almost every aspect of life, but that this can only be understood by appreciating the relationship of each entity to its dual.

What about in macroeconomics? Is there some equivalent fundamental duality? Yes. Gross Domestic Product (GDP), which we have just analyzed, is balanced by National Income (NI). Briefly, there are two sides to the GDP coin: the real side of the economy involving the production of goods and services (hence the National Product, as in equation [2.2]), and the financial side involving the financing of all economic transactions, including the payment of all wages and salaries (National Income). In dollar terms, the value of each must be one and the same so that

National Income = National Product (2.2)

where, for simplicity, we are shortening Gross Domestic Product to National Product on the right-hand side. We can thus write NI = NP in shorthand. What this says, in effect, is that every dollar that is spent for something is spent on someone. We thus have a dual representation of economic transactions. Since the two quantities must be equal, it follows that the logic utilized to forecast NP will also generate an implicit forecast of NI as well, and vice versa. Yet the two forecasting logics involved are quite different, guaranteeing that two different sets of insights will result from applying each separately. If you do apply each and the resulting forecasts are not the same, then you will be forced to make suitable revisions in each, iterating until any underlying inconsistencies are resolved. Doing so helps keep an economist honest!

This important point can be restated in the form of a useful definition: A valid forecast of GNP is one in which an analyst’s forecast of NI and NP are mutually consistent and equal.3 We are now going to utilize the logic of NI-based forecasting to provide an independent set of reasons why GDP growth will almost certainly be subpar for at least the next seven years. This should enhance the persuasiveness of our previous GDP-based forecast of a disappointing decade of growth. Note: This less familiar approach to forecasting is more demanding than the previous one. If you found my previous conclusions persuasive, you can bypass this section. But I hope you persevere. I personally favor the more rigorous insights from NI-based forecasting.

What exactly is the logic of NI forecasting? This can best be seen by rewriting the normal definition of NI in the following manner, which may not seem to be about NI at all, but is:

Government Deficit = Net Private Savings + Net Foreign Capital Inflows (2.3)

This relationship is deceptively simple. For it can be expressed in several ways. The three variables can be written in dollar terms, or in percentage of NI terms. Also, the three terms can be transposed in insightful ways, for example, Net Foreign Capital Inflows must equal the Government Deficit minus Net Private Savings. The equation as written simply says that, if the fiscal balance of a country is a fiscal deficit of, say, 10 percent of GDP (as it was in 2009), then this must and will be funded by a combination of net private savings (household and business) and net foreign capital inflow, each expressed as a share of NI, or equivalently of GDP by equation (2.2).

For example, suppose net foreign capital inflow is 4 percent of GDP. Assuming that the fiscal deficit is 10 percent of GDP, this then implies that net private savings will be 6 percent of GDP. The numbers must and do always add up. Underlying prices and quantities change in such a way that the equation always holds true. Once again, equation (2.3) is simply saying that the money to finance the government deficit must come from somewhere, meaning either from the private sector or the foreign sector or both.

Now what exactly do we mean by “Net Private Savings” and by “Net Foreign Capital Inflows”? Net Private Savings represents National Income (equivalently total private income, that is, wages and profits) minus all private spending (consumption and investment in houses and factories and new equipment). We say “all” spending here to emphasize that we are lumping together business and household sector spending.4 Net Foreign Capital Inflows is the amount of foreign savings that the nation must and does import each year to finance its trade deficit measured on a “current account” basis—the broadest measure of a trade deficit.

Most people are very confused by this reality. They understandably think that total net capital inflows (foreigners’ investments in the United States minus U.S. investments abroad) represent the net amount of capital that foreigners wish to invest in the United States. Should they become disenchanted with U.S. assets, then net capital inflows will fall and U.S. interest rates would get driven higher. But the underlying story here is much more complex, and this conclusion is completely incorrect for counterintuitive reasons.5

A Half-Century of Data

Figure 2.2 offers a graphic depiction of our National Income equation (2.3) at work during the past half-century. Note critically that, for any given year, the value of the government deficit (the heavy black line) is indeed the sum of the values of the two other lines, just as equation (2.3) requires. They all add to zero. Amusingly, in 1969, you can see that all three had values of about 0, so you do not even need to know how to add in this one case! Now focus on the far right-hand side where the data for the recession years 2007–2009 are plotted. These data portray changes in the values of two of the three variables that are astonishing in magnitude. To begin with, note that Net Private Savings lurched from –2.4 percent of GDP to +7.7 percent, an unprecedented increase of 10.1 percent. This reflected the collapse of both household and business spending relative to total National Income that is a hallmark of the most recent recession. (Keep in mind that Net Private Savings is defined as the difference between National Income and Total Private Spending.)

Figure 2.2 The Three Financial Balances (as a Percentage of Total GDP)

Note: For every year, the three data points must, and do, sum to zero.

Sources: U.S. Federal Reserve, Bureau of Economic Analysis.

In contrast, the value of the second variable (Net Foreign Capital Inflows) on the right-hand side of the equation fell slightly from 5.2 percent to 3.2 percent of GDP, reflecting an equivalent shrinkage of the U.S. trade deficit. (Remember that the net capital inflow is the mirror image of the trade deficit.) Our equation then requires that the government deficit appearing on the left must have risen by the sum of these two component changes (10.1% + (–2%)) = 8.1%. It did, of course, with the deficit soaring by an astonishing 8.1 percent, from 2.8 percent in 2007 to 10.9 percent in 2009. Once again, everything thus “adds up” in practice as it must in theory, according to the logic of equation (2.3).

Application of This Logic to Today’s Crisis

To understand how helpful the logic of NI forecasting can be, let’s apply it to today’s recovery. Suppose that the Obama administration has the following two goals. First, it wishes to increase GDP growth by 2 percent above its current 2 percent growth rate to 5 percent per annum. This is the minimum growth needed to dent the U6 unemployment rate, for reasons we have already reviewed. Second, suppose the administration seeks to reduce the fiscal deficit by 7 percent of GDP from 10 percent of GDP today back to a long-run sustainable 3 percent. Are these twin goals compatible? No. Manipulation of the equations underlying (2.3) strongly suggests that it will be almost impossible for the nation to achieve these two worthy goals. This is the problem we face.

To see why, suppose that Foreign Net Capital Inflow (equivalently the trade deficit) does not change as a percent of GDP, or at least changes little, as is widely expected. Then in order to achieve the two policy goals cited earlier, Net Private Savings must shrink by 9 percent of GDP in order for equation (2.3) to be satisfied. This is not some random opinion, but is rather a logical necessity. But can Net Private Savings fall by this magnitude? Almost certainly not. To see why, recall the definition of Net Private Savings: National Income minus total Private Expenditures. Given this definition, then in order for NPS to fall by 9 percent, then either (1) private expenditure must rise by 9 percent of GDP, or (2) if this is unchanged, then National Income (which equals GDP) must fall by 9 percent, which of course signifies a major depression! (Or there could be some mix of these two adding to the required 9 percent.) But given our twin goals of reducing the fiscal deficit by 7 percent of GDP and increasing GDP growth by at least 2 percent over what it is today, then the only way Net Private Savings could fall by 9 percent of GDP is for private spending to explode by 9 percent. Nothing else is logically possible given our goals.

But Will This Happen?

How likely is it that private spending will soar? As the data in Figure 2.2 make clear, there has only been one time when the Net Private Savings term has fallen by this magnitude due to a boom in private expenditures. This was in the second half of the 1990s during the Clinton administration. It mainly resulted from an explosion of capital spending associated with rewiring the United States for the Internet revolution. As the data in the figure make clear, no other period in modern U.S. history witnessed any comparable rise in private spending, or equivalently, any comparable decline in Net Private Savings. There is virtually no likelihood that the period 2012–2020 will witness any such boom in capital spending. Of course, business investment spending is only one form of private expenditure that could increase significantly. Other forms include household spending and residential investment. But for reasons we have already reviewed, it is highly unlikely that either of these will fill the gap.

The longer-term trade-off we are left with is unappetizing. The nation can sustain a much-higher-than-usual fiscal deficit required to sustain GDP growth—exactly what the United States has done in the past three years when it ran deficits of 10 percent of GDP that were previously unthinkable. Or it can slash the deficit bringing the rate of economic growth back to zero, or worse. Or it can adopt a halfway-house solution and reduce the deficit very modestly, the strategy it seems to be pursuing. In either of the last two cases the unemployment rate will remain unacceptably high. Note how this conclusion reached via National Income analysis reinforces our earlier forecast arrived at via more standard GDP analysis.

The benefit of utilizing the national income logic in forecasting should by now be clear. It stems from the way it enables all-important flow-of-funds constraints (equation [2.3]) to discipline armchair GDP forecasting based upon the logic of equation (2.1). When such constraints are ignored, as they usually are, the result will be incomplete and misleading GDP forecasts. The higher-level forecasting logic embedded in the National Income constraints (2.3) is indispensable in helping us identify scenarios that can occur (because they are consistent with [2.3]) versus those that cannot occur because they are inconsistent with (2.3). Some two decades ago, I had the privilege of having lunch with the late economist James Tobin of Yale. In our discussion, he stressed this point to me. He commented that many faculty members of Harvard and Yale failed to understand the importance of exploiting the NI = NP duality when forecasting the economy.

Subsequently, the late Wynne Godley of Cambridge University came to specialize in this approach to forecasting. He wrote a primer on this topic for me nine years ago, and I now realize that his highly unconventional inferences about excessive private borrowing prefigured exactly what would happen during the global financial crisis of 2007–2010. This was long before various self-styled prophets claim to have foreseen the crisis. Moreover, the depth of his logic was as arresting as the dearth of logic has been on the part of many who claim to have called the U.S. housing market collapse. I am proud to join Financial Times columnist Martin Wolf and scholars at the Jerome Levy Institute as amongst the few who have recognized the importance of Godley’s insistence on flow-of-funds analysis in valid forecasting. He deserves far more credit than he has had.

Exit from Gridlock: Mr. President, You Can Have Your Cake and Eat It, Too

Given the likelihood of subpar growth in the years ahead, of continuing high unemployment, and of increasingly vigilant bond markets, it is not surprising that a debate has been raging both in Congress and in the op-ed pages as to what government should do to prevent a Lost Decade. The two camps that have emerged consist of those who champion further stimulus spending (e.g., Paul Krugman and George Soros), even if it means larger deficits, as opposed to those who believe the deficit must be attacked right now regardless of the consequences for growth and employment (Harvard historian Niall Fergusson, British Prime Minister David Cameron, and former European Central Bank head Jean-Claude Trichet). The public is thus presented with an either/or choice between two highly problematic scenarios. Our earlier analysis is fully consistent with this depressing “either/or” conclusion.

Nonetheless, the positions staked out by both camps are wrong, as I now show. For it is possible to achieve higher growth rates of output and employment, to lower fiscal deficits placating the bond markets, as well as to achieve higher productivity growth and infrastructure improvement—all at the same time. If this Panglossian outcome seems at odds with the depressing analysis I carried out earlier, it is. Is this a paradox? No. For in my discussion above, I adopted the conventional view of what we mean by the size of the nation’s “fiscal deficit,” namely the difference between total government spending and total tax revenue. And I accepted the conventional wisdom that this must be reduced a lot. As it turns out, this concept of a deficit is highly problematic. A novel definition is needed, and will shortly be introduced. Once this is done, and a win-win policy is identified by utilizing it, then today’s Dialogue of the Deaf on the Lost Decade can be replaced by an informed consensus as to what really needs to be done and can be done. The good news is that an exit from gridlock is possible, and is needed.

Socratic Dialogue: A Rethink of Fiscal Policy

President: I have read with interest what you have said, and am keen to discover what your exit from gridlock strategy amounts to. From where I stand, matters are quite discouraging. Just as you said, the supposed experts are divided into two distinct camps, namely deficit hawks versus deficit doves. And there is no middle ground in what you properly call a Dialogue of the Deaf about deficit spending. The hawks have certainly gained the upper hand, not only in the States, but also in Europe and Australia, where deficit phobia is rampant. A dove minority, however, stresses that fiscal austerity as championed by the hawks will kill growth just as it is doing in several European economies. It argues that strong growth is the only legitimate way out of today’s morass.

I myself am a short-term dove in this sense, but there is a serious weakness in our position. We doves have tended to ignore the likelihood that bond market vigilantes will no longer tolerate the growing burden of debt that ongoing fiscal deficits imply. Given the speed and the ferocity of recent vigilante attacks in Europe, this is too important a risk for us to ignore any longer. Finally, you are spot on in pointing out that we need GDP growth much higher than today’s 2 percent rate if we are to bring total unemployment back down to normal. So what is your strategy that will permit Americans to have their cake and eat it too? And on what grounds can it be justified? No voodoo economics, I trust!

Economist: Thank you, Mr. President, for your interest. Before detailing my strategy, let me address your last point: On what grounds can it be justified? As you read earlier in Chapter 1, I champion a deductive approach to problem solving where Basic Assumptions or axioms are laid down up front, and a policy solution is then deduced from them by following the laws of logic. Assuming that parties of all stripes agree that the axioms are reasonable, then they must agree with the policy recommendation that results from the ensuing deductions.

President: Yes, I noted that point and found it both unusual and interesting. And as you said, it flies in the face of much current policy analysis where both sides stake out their positions, and then cherry-pick “facts” to back up their positions. Gridlock tends to result from the resulting shouting match. Can you deduce the solution to the Lost Decade problem in a deductive manner?

Economist: Yes. In fact I will do so in two ways. In the first instance, the deduction is quite simple, as it makes use of a redefinition of the concept of “deficit.” Once introduced, this new definition leads quite rapidly to the win-win policy solution I believe in. Importantly, I shall argue that this solution satisfies the four desiderata that everyone wants satisfied by any purported “solution” to today’s crisis: much lower fiscal deficits to placate the bond market, much higher GDP growth, much lower unemployment rates, and a solution to our infrastructure problem.

In the second instance, the approach is more abstract. Based upon extremely important research by the economists Kenneth Arrow and Mordecai Kurz at Stanford University, it is shown that a single Basic Assumption can be utilized to deduce the same policy outcome at a much deeper and more persuasive level of analysis. At a more abstract level of analysis, this second approach clarifies when and why very large government deficits are sometimes fully justified. We are precisely at such a stage now. Given limitations of space, and the somewhat technical nature of the required deductions, I will simply sketch the argument graphically in the last part of the chapter, stressing its relevance and indeed its originality.

President: Good. Having two altogether different sources of support will be helpful if you really have deduced a solution that purportedly will satisfy most everyone. It will be that much easier to sell to Congress. Now let’s cut to the chase. What is your policy solution, and how do you justify it via these two lines of argument?

Economist: The solution lies in redefining the very concept of a “deficit” in a manner that permits a new theory of “deficit spending,” a theory that can be applied to the situation in the United States right now. When I say “new” here, let me be clear that the basic idea is not so much new as it is currently unrecognized. Out of some 500 op-ed columns on the subject of deficit spending that I have read in the past few years, I have seen perhaps four articles that set forth the basic thrust of this approach. Yet none of these four suggested that my approach to deficit spending could simultaneously satisfy all four desiderata that must be met to avoid a Lost Decade. Nor have I ever seen this approach justified by first principles stemming from the Arrow-Kurz theory that lies at the foundation of fiscal and macroeconomic theory. Given the urgency of the U.S. crisis, and given today’s strong prejudices against ongoing large fiscal deficits, a very strong justification is needed for what I am going to propose. This will be provided.

President: Can you explain the main point here in the simplest possible terms?

“Good” versus “Bad” Deficits, and the Main Policy Proposal

Economist: Yes. Please consider Figure 2.3, which contrasts the fiscal status of two countries with ostensibly identical deficits. Assume that Country A’s government spends $4 trillion on defense, administrative costs, interest expenses on its debt, and transfer payments such as Social Security and Medicare. Its tax revenues of $3 trillion fall $1 trillion short of this $4 trillion, so that the nation runs a deficit of $1 trillion. More specifically, its treasury department will have to issue $1 trillion in new government bonds. The expense of servicing this new debt (or repaying it) will fall on tomorrow’s taxpayers, who just might renege on it if it gets too large.

Figure 2.3 How to Fix the U.S. Economy and Labor Market While Not Upsetting the Bond Market Vigilantes

The Moral: It is the composition and quality of total government spending that matters, not the “size of the deficit.”

As the magnitude of total national debt outstanding grows each year because of these marginal additions to total debt (i.e., each year’s deficit), a point will be reached where the bond market fears future insolvency, or else a printing away of the debt. As a result, investors will demand higher yields, which crimp the growth rate of the economy, and cause the cost of refinancing the debt to explode. The infamous “debt trap” could soon be reached—a fiscal red hole from which few nations ever escape. Country A is modeled after the United States, of course, and during 2010–2011 investors worldwide began to question its long-term solvency, especially in light of the inability of both political parties to cope with runaway spending.

President: But what is different about Country B, whose revenues and outlays are identical to those of country A, but which you claim has no deficit? Is this a trick? Are you the Houdini of macroeconomics?

Economist: Thank you, but no. Country B differs in one regard: Of its total $4 trillion in spending, $1 trillion is spent on profitable investments (human capital and infrastructure investments) which are certified by an independent research organization to generate a positive expected return on capital, as calculated within the venerable field of public finance. The remaining $3 trillion are expenditures on defense, interest payments, administrative costs, and transfer payments like Social Security and Medicare. When bond market vigilantes understand this breakout, they see a nation whose necessary if unproductive spending of $3 trillion is fully matched by current tax revenue. So there is no deficit from unproductive spending.

On the other hand, the nation is borrowing an extra trillion dollars to invest in projects that are “certifiably productive,” as it were, and that pay for themselves over time in the same way as any productive capital investments in the private sector pay off. So, overall, no new debt has been chalked up that future generations must service. That is why the figure shows a deficit of zero. The bond markets are placated, and interest rates are not driven up.

President: I am supposed to think of Country A as America today, and Country B as America the way it could be tomorrow, correct?

Economist: Yes. More specifically, think of Country B as the America that would be, were Congress able to redirect a good chunk of government spending (up to $1 trillion per year) away from existing spending toward productive spending. Historically, such spending included investments in the highway system, the space program, the Internet (known originally as the DARPANET), the interstate highway system, R&D, the energy grid, water resources, and so forth. Congress is not to cut $1 trillion as many deficit hawks would like. Rather, it is to reconfigure total spending so as to boost GDP, jobs, and productivity. There are no net layoffs, and no fiscal drag at all.

President: So your sleight of hand is to introduce two kinds of deficits, “good” and “bad” deficits as it were, that differ according to the type of spending that generated the deficit.

Economist: Yes. Emphasis accordingly must shift away from the overall size of the deficit to its composition of good versus bad spending. I am not claiming here that nonproductive spending is bad per se, as it is not. But it is spending that must be matched by tax revenues so as to no longer increase the burden on future Americans.

President: The principle here is straightforward, and come to think of it, in the private sector, no corporation would claim ongoing losses by “expensing” investment spending. Rather, the firm capitalizes the investment expense, and only deducts from income the amortized portion of the investment each year. In doing this, there is no “deficit” due to investment spending.

Economist: Yes, indeed, and it has long been recognized that governments should develop an accounting system that partitions its own expenditures analogously. Some governments have attempted to do so in the past, but this was neither expected nor required. Today, however, it would be wise for governments to develop a proper sets of books. For in coming years, transparency of this kind will be important for keeping the bond market at bay.

How the Proposed Policy Satisfies the Four Desiderata

President: As you doubtless suspect, I have a lot of questions about what you are proposing, although it is difficult to disagree with your basic logic. But before we pass from fantasy to reality, and I grill you on some of the details of your plan, please explain how this strategy would satisfy all four criteria that you cited previously: more rapid GDP growth, reduced unemployment, a contented bond market, and a solution to our infrastructure crisis. I do not understand this claim.

Economist: First consider GDP growth. Under the proposed strategy, GDP growth would be much greater than otherwise. To begin with, there would be no fiscal contraction of the kind we are already beginning to experience. The nature of the deficit would change, but not its size. Usually, when the IMF orders a debtor nation to slash its fiscal deficit from, say, 10 percent of GDP to 3 percent, the nation suffers a 10 percent – 3 percent = 7 percent loss of GDP, other things being equal. We saw why in equation (2.1), which defined GDP. Of course, this loss is often spread out over several years, lessening the pain.

President: Is GDP positively impacted in any other way by your plan?

Economist: Yes. The nature of the new spending—on investment rather than transfer payments—triggers all those “multiplier and accelerator” effects we learned about in Econ 101. These factors raise economic growth even more. Recall how, if funds are given to a master design firm to create a new airport, or whatever, this firm in turn will borrow against these funds and hire a large contractor who in turn will borrow and hire 15 subcontractors, and so forth. The jobs multiply, sometimes at an accelerating rate. This multiplier is estimated to be approximately 2.5 versus 1 when the money is spent on a transfer payment for keeping a given worker at his existing job.

There is yet another boost to GDP aside from the jobs multiplier story. When money is spent on profitable projects only—think bridges to somewhere rather than bridges to nowhere—productivity growth will rise with every dollar spent. And higher productivity growth implies higher GDP growth in the longer run.

President: Please elaborate.

Economist: When a nation lets its infrastructure run down, as the United States has, then the deterioration can reach a critical point. Bridges start collapsing much more frequently. The national electric grid is stressed out and falters. Blackouts become much more frequent. Old refineries experience safety and maintenance problems. Nuclear plants become much more hazardous. Traffic lines lengthen exponentially. As this happens, investment spending of the right kind can significantly boost productivity growth—much more than in normal circumstances—since it will both prevent systemic collapse and will permit an improvement over the status quo prior to collapse. This is very important since the United States may well be reaching a critical point of infrastructure collapse during the current decade. The nation simply must address its infrastructure crisis on a truly grand scale, and stop equating infrastructure investment with filling potholes.

President: So all of this propels GDP growth even higher. Growth benefits from a lack of fiscal drag, from the multiplier effect, and from higher productivity. What about job growth per se?

Economist: Job creation will accelerate much more rapidly under my plan than it otherwise would. First, faster GDP growth from the lack of any fiscal drag will itself create jobs. Remember, there will be no reduction of overall fiscal stimulus at all—just a reconfiguration of spending toward investment. Secondly, the shift of expenditures from transfer payment spending, which has zero job-creation effects, to investment spending with its knock-on accelerator/multiplier employment effects will cause a much greater rate of job creation than otherwise, as was just noted. Both of these points should be central to any strategy for accelerated job creation.

President: Shifting to another one of your four objectives, what will the bond market say about all this? Will it finally be satisfied that we have stopped mortgaging our future?

Economist: Most probably, provided you and the Congress make clear that you truly will rack up “good” deficits as needed, and bring to an end the buildup of bad debt that our children may one day default on, and that rattles the bond market. You will also have to persuade the bond market that your infrastructure program will restrict itself to investments that earn a positive expected rate of return on the funds borrowed to pay for them. I will discuss later exactly what that means in reality.

Incidentally, assuming that such investments will earn an expected profit, you have the option of funding them not merely by government borrowing (thus adding to the deficit), but also by raising private capital, possibly backed by a government guarantee. The ways in which governments can fund projects have proliferated in recent decades. By approaching bankers at the Macquarie Bank in Australia, which has specialized in such matters, you will be surprised by how many ways there are to skin this cat. You can also approach the governments of the States of Victoria in Australia, of Ontario in Canada, and the nations of Chile and Turkey to discover still other forms of creative government financing. Once a case can be made that the projects are needed and thus profitable in a sense which will be defined further on in this dialogue, then all kinds of financing opportunities will present themselves.

Finally, since the case for running a significant “good” deficit well into a business cycle recovery has not been made before, at least not in the political arena, it would help if you would direct the OMB to prepare a new set of government accounts that separate out capital investment spending from normal recurring expenditures. With these accounts in hand, you could demonstrate to an otherwise skeptical public and bond market how the “bad” U.S. fiscal deficit properly computed could drop to zero under the proposed policies over the next seven years or so. Once again, this deficit is the difference between unproductive (but necessary) government spending and total tax receipts. Showing this would help you win over deficit hawks and other naysayers in today’s environment of deficit phobia and austerity.

President: Given today’s mindset, this will be quite a challenge. But I like the way your proposal seems like Common Sense 101, to utilize one of your favorite phrases. You have put your case well, as far as it goes. Now, however, let me express some serious concerns. First, what you are proposing is quite radical, even if it is commonsensical. You are suggesting that those on the Left like Paul Krugman and George Soros are right in arguing that we should not significantly cut our fiscal deficit of 10 percent of GDP at all. They are dead set against any fiscal contraction that might further increase unemployment. But so are you. Where you seem to differ from them is your insistence that the composition of government spending must be changed, tilting it toward profitable investment spending on a veritable Marshall Plan scale, and away from unproductive transfer payments. Am I correct here?

Economist: Yes, and I find it extraordinary that in all their op-ed pieces, the two liberals you cite have not raised the all-important distinction between “good” and “bad” deficits as defined earlier. This should be the central point in any rethink of fiscal policy today.

President: Can the bold plan you are proposing be justified on any further grounds? A proposal this radical will need all the support it can get given today’s deficit phobia.

A Further Justification of the Proposed Marshall Plan

Economist: Yes. To begin with, I have already given you the principal justification for the proposal: It satisfies the four national policy objectives stressed earlier. Had I gone a bit further, I could also have demonstrated that it is the only strategy that can do so given the very tight constraints that these objectives impose on U.S. policy makers. But let me now take a different tack and clarify why a wholly new strategy is called for at this juncture in U.S. history, due to a novel set of circumstances that have arisen and that will constrain your policy choices.

President: Is there in fact something really new about today’s circumstances? Some dispute the claim that this time really is different.

Economist: Indeed there is. To begin with, policy makers must attempt to achieve much higher growth after failing to do so despite the largest monetary and fiscal stimulus in over five decades. This challenge truly is a post-war first! The level of distress in the labor and housing markets and in the public sector is also highly unusual. So is the extent of sluggish demand in the private sector. Finally, there is the growing likelihood that our aging infrastructure could go critical by the end of the decade. This is also new.

But what is completely new is that the United States for the first time in a century has outlived its welcome in the global bond market at the very time it will need ongoing fiscal stimulus to perform the heavy lifting of its recovery. Monetary policy can do little more than it has. Demand in the private sector is flat. Therefore only public sector borrowing and spending can do the job. Yet bond market vigilantes could soon be up in arms. At this writing, rating agencies have put the U.S. Treasury on their watch list. Higher interest rates and soaring debt refinancing costs could well be the result, just as occurred in several European countries during 2010–2011. The fact that U.S. yields recently have been so low should deceive no one about the future. The yields on many Euro-bonds were low too, until the firestorms of 2010 and 2011. Finally, the continuing fall of the dollar underscores how much confidence in the United States has already been lost overseas.

All in all, this set of new circumstances explains why a radically new strategy is called for, and the one I have sketched out is designed to accommodate all these realities. To state it differently, the proposed policy is consistent with all these constraints, unlike any other recovery strategy proposed to date. My strategy has the potential of transforming a Lost Decade into a Winning Decade in which the United States finally gets down to business, attacks the right problems in the right way, and satisfies all four of its principal policy goals in one fell swoop.

Three Potential Difficulties

President: I am with you thus far. We are indeed in a new situation, bound by new constraints, and a novel policy is in order. Nonetheless, I have some serious concerns about your own plan. First, does the nation really need the vast scale of public sector infrastructure you are calling for, up to $1 trillion per year over a decade, or a $10 trillion investment? This is a much larger figure than is usually cited. Second, is such a plan physically feasible? Do we have the nonfinancial resources to carry it out? Third, how can I convince investors that infrastructure projects once identified really are “profitable” and earn a positive rate of return on invested capital? As you know, the public is very cynical, and doubts that the government can do anything right. Boston’s notorious “Big Dig” overrun is now seen as the rule, not the exception, of government projects.

Economist: Let me sketch answers to all three of your questions, each of which represents a legitimate concern. As to your first question regarding the scale of the investment plan that is needed, there are two considerations. First, how much infrastructure of various kinds would ideally be needed were the nation to decide to leave to its children and grandchildren top-notch infrastructure they can be proud of? As far as I know, this question has neither been asked nor answered. Rather, we hear piecemeal accounts of how much is needed to prevent bridges from falling down, or to fill potholes, or whatever. Second, how much public sector stimulus is needed to get the nation back on track with much faster growth and much lower unemployment, while placating the bond market? The answers to these two questions are critically interrelated in what I have proposed.

Ideally, the numerical answer to the first question would be the same as to the second! I will argue that this is precisely the case in the United States today. Up to an approximation, the nation needs $1 trillion of both government-initiated investment and of ongoing fiscal stimulus on an annual basis. This “fit” very much works in our favor, and this is fortuitous since there need be no such fit at all. For example, consider the case of Japan. On the one hand, Japan has little need of infrastructure spending due to excellent maintenance and ongoing investment over the decades. On the other hand, Japan needs ongoing fiscal stimulus because of its well-known stagnation.

Given these realities, my plan would not work in Japan: It already has the infrastructure it needs, so that further investment along the lines I have proposed would amount to building “roads to nowhere,” so to speak. This is a code word for “roads that generate a negative expected return on capital invested.” Any such projects would thus violate my plan’s requirement that all investments be profitable, and not pose a burden on the grandchildren that upsets the bond market. Only “roads to somewhere” are allowed in my plan.

President: A combination of circumstances opposite to those of Japan would be those of a country that has a great need for deferred infrastructure investment spending, but whose economy is roaring along just fine so that no fiscal stimulus is needed at all. Is that correct?

Economist: Exactly. The question for the United States is thus whether we need both massive infrastructure spending and significant ongoing fiscal stimulus. Based upon the earlier analysis, my conclusion is that we do need both. I suspect you agree. As regards the need for infrastructure, the United States and the United Kingdom rank at the top of most lists as having the most dilapidated stock of public sector capital out of all the OECD nations. Neither nation has put a new roof on its house in half a century. Let me reemphasize that what is at stake is not merely the repair of deteriorating public sector capital stock, but the development of completely new capital stock of higher quality than that it replaces. For example, while we need to fix older roads and bridges and tunnels, we also need new high-speed trains, new energy grids, new oil refineries, and so on.

President: And don’t forget the human capital investment that you mentioned earlier, if you really are shooting for the moon.

Economist: Yes, investment targeted toward jacking up productivity in both K–12 education and health-care infrastructure (see Chapter 3) will prove very important. To cut to the chase, when you compile a wish list of total public investment required to upgrade the entire infrastructure of the nation over the next decade—not just roads and bridges—then the sum lies between $10 trillion and possibly even $15 trillion according to many discussions I have had. For example, the National Academy of Engineers has estimated that over $4 trillion will be needed simply to repair and upgrade all of the nation’s bridges and roads. That is before attending to any of the other kinds of infrastructure that are needed.

This is a very large country and it all adds up to a very large number when a decade-long perspective is adopted, or better a two-decade perspective were we truly serious. But what a moment: Is $10 trillion of new capital stock that large a number in a nation whose GDP is $15 trillion annually, and that has irresponsibly deferred infrastructure maintenance for five decades? The more I research and ask questions, the more this appears to be not a large amount at all. Perhaps $20 trillion would better approximate the ideal level of investment over, say, two or even three decades.

President: What you are calling for is a veritable Marshall Plan dedicated to infrastructure, in the broadest sense of this term. And if the figure really does amount to $10 or $15 trillion over a decade or so, then this would amount to at least $1 trillion of such spending a year.

Economist: Yes, and note that $1 trillion per year equals the size of the ongoing fiscal stimulus needed to prevent a Lost Decade from materializing. This is why I asserted that, in the case of the United States at present, there is a nearly perfect fit between the size of investment spending required and the size of fiscal stimulus needed throughout the decade, assuming that private sector spending continues to be inadequate for the level of growth needed to redress the unemployment crisis.

Going one step deeper, the payoff from the proposed strategy is even more compelling than all this suggests. As you know, the United States has been widely criticized for underinvesting and overconsuming for several decades. We eclipse all other nations in spending a whopping 71 percent of GDP on consumption. The result of this imbalance has been exactly what the late Harvard economist John Kenneth Galbraith predicted long ago in his book The Affluent Society: private splendor, and public squalor. The proposed plan would reverse this situation, tilting the consumption share of GDP back to a more balanced 65 percent of GDP within a decade or so, while radically increasing the investment share.

Are the Needed Resources Available?

President: Good. My second question is whether we have the nonfinancial resources to undertake such a plan. For example, where would the manpower come from? I am taking your word that financial resources would be available provided there is good evidence that the spending will be productive.

Economist: Remarkably, we probably can garner the resources required. To begin with, the collapse of the housing industry saw 2,003,000 construction workers and engineers lose their jobs. As of mid-2011 there are still 1,137,000 construction workers looking for work. Additional slack in the system comes from today’s relatively low level of corporate investment spending, and the resources this frees up. There is also the large army of unemployed workers outside of construction—many with skills—that have not yet been able to secure employment. Finally, there would and should be a reallocation of existing workers from sectors in trouble, including state and local governments, toward design, construction, engineering, and project management. As for new workforce entrants, talented students will pick up on this development and reorient their careers away from Facebook and finance and toward engineering and “making things.” It is high time.

All in all, I do believe we have the resources needed to carry out the required investments. But look at matters in reverse. A belief that the required resources might not be available is probably mistaken. In a market economy governed by a price system, prices and quantities miraculously adjust so as to make resources available where they are needed, and they do so with an efficiency that has proven remarkable time after time. If opportunities to make a living shift toward infrastructure, you can be sure that the resources will follow.

President: Okay, you may be right that your policy is feasible. But is it politically feasible? For example, your plan requires a gradual shift of government spending out of sectors dependent upon transfer payments and into sectors supporting very large new infrastructure spending. That will be a hard sell, especially amongst Democrats.

Economist: Yes, some Democratic Party spokesmen will probably brand the idea as “heartless,” but it should be easy for you to refute such claims, especially if you truly take charge and lead, which is what people want. You have two aces in your political deck. First, this plan will create more jobs than any other plan can, and politically, joblessness has become Issue Number 1 in this nation. Second, you must make clear this is not some zero-sum game where for every infrastructure job created, another job must be lost. Due to the job multiplier effect, there will be a significant net increase in jobs. Furthermore, to the extent that reduced transfer payments do adversely impact certain beneficiaries, it’s time to remind such workers that the other shoe sometimes must fall in life. Your original stimulus program protected many state and local workers’ jobs, whereas no protection was afforded to those in construction. It may well be time for a flip-flop here in the name of net job creation, as well as in the name of fairness.

Legal System Difficulties—NIMBYism in Particular

President: There is one final aspect of feasibility that worries me and that you have failed to mention. This concerns the role of the U.S. legal system in implementing your plan. Assuming you are correct that we have the labor and capital required for the proposed investments, will our legal system permit the plan to be realized? Or will it be stymied by NIMBYism—that “not-in-my-back-yard” form of project paralysis? Will well-intended projects languish in courts during years of lawsuits, thereby discrediting the entire program and driving returns negative?

Economist: You are now putting your finger on the obstacle that bothers me the most. This really is a sink-or-swim issue. I have discussed this matter with lawyers and public policy analysts. The root question seems to be whether the federal government on its own has the authority and the power to push through this kind of an investment agenda. No one has given me a convincing answer. However, there is solace in the example of the construction of the vast Interstate Highway System over the decades following World War II. President Eisenhower utilized national security arguments to insist upon construction. I myself can recall those large ICBM missiles being shipped around the country to widely separated missile silos. Of course, this was during the height of the Cold War, and the U.S. landscape was much less developed than it is now.

President: All this is true, but the reality is that today’s legal system is much more complex, and specialized interest groups are much more embedded than they used to be. For example, there was no firmly entrenched environmental lobby to prevent the construction of Route 80 across the country lest the habitat of some rare brown-spotted squirrel be disturbed.

Economist: This is true. Nonetheless, you should make it a priority to gather the best legal information you can on this issue. The role of the Supreme Court will be very important. They recently voted for a significant expansion of the right of eminent domain whereby private property can be seized on behalf of private sector commercial interests, provided that the larger community will benefit from job creation and other “externalities” resulting from development.

President: Yes, and this was a very controversial verdict. Nonetheless, it hints at ways your program could succeed. Perhaps the twin needs for job creation today and better infrastructure for future generations would provide a legal basis for a much broader federal mandate than has hitherto existed. This is well worth thinking about.

Economist: There is a further argument that bolsters expanded eminent domain in the public sector. It is easy to think of infrastructure improvement as a luxury—one to which the nation can perfectly well say “No.” We have done so for decades. But it is not, in fact, a luxury. It is fast becoming a necessity for the simple reason that infrastructure deterioration will begin to go critical, thereby threatening the public’s safety and welfare. On these grounds, I can foresee a day when the Supreme Court steps in and mandates infrastructure spending for reasons of public safety. This might happen after hundreds of people are hurled to their death after 20 more bridges collapse, or after a series of severe brownouts drive up unemployment and cause hundreds of deaths in hospitals.

President: You are being a bit sloppy here. Do not forget that the Supreme Court does not mandate anything. It merely arrives at verdicts. What would happen is that Congress in Washington would pass a law ceding vast new powers to the federal government in the interest of “timely, necessary infrastructure investment and maintenance.” States’ rights lawyers would presumably challenge such a law in the Supreme Court, and the court in turn could rule on behalf of the federal government. This is perhaps the best hope.

Economist: You are correct. One other improvement needs to be made in the battle against NIMBYism. Game theory teaches us the importance of awarding generous side-payments to losers in breaching bargaining impasses. Such payments are often the best way to transform a zero-sum game into a positive sum or win-win game. I have always thought that we are not creative in designing them, and do not utilize them enough. This is especially true in the case of contending with NIMBYism.

The Identification and Ranking of Public Investment Projects by Their Return

President: I have one final concern about your plan. How will we identify those profitable investment projects that should be undertaken? How do we assure ourselves that we will end up with high return bridges to somewhere as opposed to boondoggles to nowhere? As I already said, almost everyone is cynical about the government’s ability to do anything right anymore.

Economist: I am glad you are pressing me on this issue, as it lies at the heart of my proposals. I have argued for a truly large-scale investment, a domestic Marshall Plan as you put it. We need this not only because ongoing fiscal stimulus is required to solve the four problems the nation must solve, but also because we owe future generations a capital stock superior to what we have now.

To support this effort, there must be a large, new public/private Infrastructure Bank with significant foreign support. Just as the Old World helped finance the New World when English and German banks helped finance the U.S. canal systems and railroads of the nineteenth century, the New World now can help the Old World rebuild itself with their surplus savings—providing of course that returns from doing so are attractive.

President: Several people have proposed infrastructure banks since I took office. Executives at the Fluor Corporation are very keen on this, I believe, and they are not alone. So are many politicians, including myself. In a related direction, the eminent New York financier Felix Rohatyn wrote an excellent book a few years ago about the proper role of government in financing infrastructure, Bold Endeavors.6 He revealed the indispensable role of government in creating the infrastructure of the nation during the past two centuries. But what you are proposing sounds somewhat different.

Incidentally, does the United States actually need foreign investment? I thought your proposal would be financed domestically.

Economist: We do and we do not need their capital. This is tricky. To begin with, since we run a large trade deficit of about $500 billion, we automatically receive large net foreign capital inflows of a corresponding magnitude. You will recall from the flow of funds discussion earlier that net foreign capital inflow equals and finances the trade deficit (equation [2.3]). What matters is whether these funds flow into purchases of more T-bills by foreigners, or into more productive investments. I think we would benefit politically by having investors like the sovereign wealth funds of Singapore, Norway, and China represented on the board of the proposed infrastructure bank. Their presence and possibly their capital would reduce the temptation of politicians in Washington to meddle in the selection of investment projects. But in reality, foreigners can perfectly well go on investing in T-bills, and we would then finance the program domestically.

When you ask whether my proposal is different from Rohatyn’s, the answer is yes. It is different partly because the scope of what I propose is so large, and partly because of the need to focus on the incentive structure and politics of the new bank and its activities. Three main questions arise here. First, how do you identify at an analytical level which projects should be funded in which order? Second, what incentive structure will prevent “politics” from undermining the program’s integrity? And third, what kinds of mixed public/private investment structures should be utilized?

As I am not going to expand further upon this last point, it is simply worth noting that modern finance and investment banking have developed a remarkable tool kit for helping governments optimally finance worthy projects. As mentioned earlier, the contributions of Macquarie Bank and others have shown that there will be some blending and sequencing of private and public financing that will be optimal for any given project. The United States is a complete laggard here, and we could learn a lot from the experiences of many foreign countries and their investment bankers.

President: What about my final concern, identifying and ranking potential projects? Can this be made nonpolitical?

Economist: The new infrastructure bank would hire several hundred very well-trained and very well-paid analysts skilled in both private sector and public sector investment evaluation. Their job would be to evaluate all potential projects from a master list, and to then quantify and rank their risk-adjusted expected returns in order from highest to lowest. There could be 22 candidate high-speed rail systems. Some would end up with returns of 14 percent, others with –9 percent. With such a ranking in hand, investments would then be made in the order of the highest return to the lowest return—where the cutoff point would be the appropriate benchmark market rate of return. Precisely how to make such assessments is discussed next.

President: But how can this process avoid being politicized? Is there any objective way to estimate expected returns on capital invested in a new bullet train or highway system?

An Objective Assessment of Returns

Economist: The good news on this front is that, during the past 50 years, we really have learned how to assess the expected risk-adjusted returns from both private and public investments. And these two kinds of returns must be calculated very differently. As for ranking public-sector investments, recent analytical discoveries have in effect breathed new life into the time-honored field known as Public Finance. Many of these issues were addressed in a fully unified manner in one of the landmark books in the history of macroeconomics, a 1970 book entitled Public Investment, The Rate of Return, and Optimal Fiscal Policy. As is noted later, this treatise unified four different branches of economics into a whole, and addressed many of the concerns you have cited. It was coauthored by the economists Kenneth Arrow and Mordecai Kurz of Stanford University, and will be discussed at length further on.

President: What exactly is it that makes public investments much harder to evaluate than private investments?

Economist: It is the issue of “externalities,” otherwise known as “nonmarket effects” or “spillover effects” that make public sector investment analysis more difficult. If you and I invest in an ice cream parlor, our goal is to maximize returns from the production and sale of ice cream. Returns are easily measured. An observer just counts the dollars in versus the dollars out to arrive at net profit. The discounted stream of future profits divided by the capital needed to create and open the shop is the rate of return on investment.

President: Whereas the decision to create the Interstate Highway System had spillover effects that were very difficult to forecast and calculate. Is that it?

Economist: Basically, yes. The difficulty does not center on projecting total costs, but rather on projecting total benefits—and note that I do not say revenues. To be sure, if you build tollbooths every 50 miles of highway, there will be unambiguous benefits in the form of revenues. But above and beyond these, there will be the government tax revenues earned on the huge increase in GDP generated over the decades as new cities and suburbs grow up around the highway system. An Interstate is merely an artery, but commerce gets carried out in the capillaries and veins that end up surrounding it. Computations have shown that this enormous infrastructure project during the decades after World War II served to create a New America, with large investment returns to boot. The same was true of the creation of the Internet, a very low-cost spinoff of ARPA research in the 1960s, and originally known as the DARPANET. The Rohatyn book cited earlier sketches the large payoffs historically reaped from mega-investments, and does so in a very accessible manner.

President: Are you saying that somewhat intangible benefits like these can be now be modeled and quantified like return on investment (ROI) in the private sector?

Economist: Absolutely. Let me give you another example. The GDP of India would increase considerably should India gain the infrastructure needed to reduce commuting time in half. Remember that GDP is a pie that is baked in offices and in factories, not on overcrowded and delay-plagued train platforms. Government will receive a large ongoing tax dividend from the new income made possible by better Indian rail service, a figure estimated to be for larger than the cost of the required infrastructure. And this measure of return does not include revenues from selling tickets.

President: What about Amtrak rail service in the northeast U.S. corridor? People always complain that it loses money.

Economist: Their complaints are largely misplaced. Ticket sale revenues were never intended to measure the total profit or return on a public service like Amtrak between Washington, DC, and Boston. These trains permit millions of people a year to get where they are going fast, and to work more, regardless of weather conditions. The government receives tax revenues on the extra GDP realized.

President: You will agree that the risk in estimating both the costs and benefits of public projects will be very large. Doesn’t this risk have to be significantly discounted, thereby lowering the expected risk-adjusted returns of the project?

Economist: No. The economist Kenneth Arrow showed long ago that, because of the government’s ability to widely diversify risks across its huge portfolio of investments, the correct “social” risk premium in this context is in fact zero. This boosts returns on public investment above returns on comparable private investment (other things being equal) since significant risk premiums are appropriate in the private sector.

President: But you agree that there will be large risks?

Economist: Yes, as there are whenever Johnson & Johnson decide to introduce a new product line. Just ask their R&D directors how risky this can be! There is always large risk in any project. The problem is that public sector officials to date have had less of an incentive to assess and manage their risks than have their counterparts in the private sector. But there is a much more important point lurking here. Since when should America turn its back on risk? The answer is, never.

Keeping Analysts Honest in the New Bank

President: The second question I have concerns the politicization of the ranking of projects that you expect your highly qualified analysts to generate. What will keep them and the system honest?

Economist: To begin with, the analysts hired by the new bank are to be very highly paid, with their performance fully dependent upon transparency and objectivity of their analyses. Their high pay reflects the “Singapore Model” wherein government officials of importance are paid extremely well so as to avoid the temptation of corruption. Second, the results of their analyses will be public, transparent, and available online, with outside parties available to contribute their concerns and suggestions. Third, foreigners should and probably will be involved both as investors and board members. Powerful Congressional representatives hoping to get “their” negative return projects approved on the basis of their political clout will have trouble explaining their reasoning to those Norwegian, Singaporean, and Chinese investors sitting on the infrastructure bank’s board!

President: It is said that the advent of cost/benefit analysis in the 1960s created more honest politicians in Washington than ever before. Officials finally had to justify the costs of projects they favored. Are you extending this logic?

Economist: Yes, and to a much higher degree. Historically, the same thing happened with the invention of double-entry bookkeeping in Genoa in the fifteenth century: For the first time investors could actually learn where their money was going! It became harder to steal it from them. All in all, I would like to see the large management consultancies and investment banks incentivized to support this new infrastructure bank. As for young people interested in finance, as so many now are, this could become a “cool” place to work, a high-profile career opportunity, much as the Peace Corps became under Presidents Kennedy and Johnson.

Need for Powerful Leadership

Economist: There is one final area where the ball lies in your court, Mr. President. This concerns leadership. You will have to endorse and sell this proposal to the people—not just to Congress—assuming you believe in it. In doing so, you must talk up to the people. Do not just offer “hope,” your campaign theme, but concrete hope showing that there really is a pot of gold at the end of the rainbow, that is, that concrete problems can be solved, and the American dream restored. Many Americans now believe that today’s problems cannot be solved given today’s contentious gridlock. You might attempt to restore optimism in the following two-step manner:

Step 1: Explain that the nation must solve all four challenging problems identified earlier if this is not to be a Lost Decade, and if we are to bequeath to our children a better future. Spell out all four challenges at every opportunity: the need to create jobs, to raise the rate of GDP growth, to redress the infrastructure scandal, and to avoid increasing the nation’s stock of “bad” debt so as to keep global bond markets at bay.

Step 2: Demonstrate that there is one and only one solution that solves all four problems at once. Don’t just state this. Demonstrate it aggressively, with the use of arresting visual aids of the kind presidential candidate Ross Perot once successfully used. And having won the people over, embarrass the members of Congress if they try to play partisan politics and derail your plan. The Senators of ancient Rome were supposedly more frightened by the mob that the emperor could stir up than by the emperor himself. So stir up the beast to demand action. Do not leave it to Congress. With all due respect, doing so has been your biggest mistake to date.

To win over sufficient votes in Congress, you should exploit two political opportunities that bolster your case. First, the proposed strategy is win-win in nature and should be able to be supported by both sides of the aisle. The Left can claim that this is Big Government coming to the rescue of the nation as in the 1930s. The Right for the first time can claim that resources will be allocated nonwastefully by a new independent bank, in strict accord with capitalism’s insistence upon a proper return to capital. Second, you should remind members of Congress that, in forthcoming elections, their refusal to support the legislation is equivalent to opposing the electoral mantra of “jobs, jobs, and jobs.” By the time of the election dates of 2012 and 2016, opposing large-scale job creation will not go over well with voters.

President: Good observations. And yes, there really is a lot of common ground to unite both sides of the aisle behind your proposal. And as you keep stressing, there really is nothing either Left wing or Right wing about it.

In concluding this discussion, I want to go back to an intriguing claim you made early on. You said there are two very different ways in which you can deduce your proposed strategy from first principles. What is the second?

Economist: The second is altogether different. But first, be sure you understand how the earlier discussion proceeded from a set of Basic Assumptions. The type of deductive logic used was to list four principal policy goals, and then to determine whether a policy exists that satisfies all four. I have done that, and shown that such a policy exists. With a more formal macroeconomic model at hand, this result could be strengthened to show that the proposed solution is the only solution that can satisfy these particular criteria.

A Deeper Approach to Justifying the Proposal

President: What then is the second approach?

Economist: To understand the second way in which the proposed policy uniquely satisfies a few very basic axioms, advanced macroeconomic theory is needed, in particular the deductive logic utilized in the Arrow-Kurz book cited earlier. All I am going to do is to sketch the nature of the argument here, and to do so at a qualitative, not a quantitative level of analysis. This will suffice to express the main ideas, ideas that are very interesting indeed.

President: Can you please express these core ideas very simply?

Economist: Yes. Suppose you want to fly yourself and a friend to the moon and back in the shortest time possible. You have a good rocket to get you there. What do you do? First, you have to determine the best trajectory to get you there and back, where by “best” I mean “most fuel-efficient” trajectory. This is not easy, or course, since there are an infinite number of trajectories, each requiring a different amount of time and fuel. Second, you must figure out how to adjust the rocket’s thrusters along the way so that the rocket sticks to the optimal trajectory once it has been identified.

President: What does this have to do with anything?

Economist: Hang on, please. Now consider an economic Philosopher King who views running his economy in a similar fashion. His goal is to target the maximal well-being of the citizenry (suitably defined), and to figure out how to use fiscal and monetary thrusters to help achieve maximal well-being. In particular, he must determine the size of deficits and surpluses in both the private and public sector, as well as the mix of private sector and public sector investments that maximizes well-being. What should the Philosopher King do? What policies will be optimal, given his goal?7

President: So optimally controlling a rocket ship to minimize fuel consumption is somehow analogous to my efforts to set economic policy on a course that avoids a Lost Decade and maximizes happiness? And you will now tell me that your main policy proposal is the only one that maximizes the well-being of the citizenry. Is that correct?

Economist: Yes. The principle basic assumption this time, as you suggest, is the maximization of well-being—a much more abstract axiom than concrete policy goals like reducing unemployment, raising growth, shrinking the deficit, and so forth. Moreover, you might be surprised to know that the analytical model required to solve the Philosopher King’s problem is almost identical to that required by the captain of the rocket ship to determine a least-fuel trajectory.8

President: Is the concept of “maximizing the well-being of the people” well-defined, or is it a bit fuzzy?

Economist: One of the great successes in economics and philosophy during the past seven decades has been the clarification of exactly what this means, and even of how to measure it. The concept originated amongst the early British utilitarians such as Jeremy Bentham, who proposed that government policies should be chosen so as to “maximize the greatest good of the greatest number.” In more modern guise, this policy objective becomes, “maximize the aggregate satisfaction of all the people,” where each person’s “satisfaction” is measured by that person’s own intensity of preferences for the alternative outcomes at stake.9

President: So what exactly is the link between this highly abstract objective and the policy proposals you have set forth? Does maximizing the greatest good truly imply your jobs-creating infrastructure investment plan, thereby justifying it from first principles, as you put it?

Economist: Yes it does, assuming a host of subsidiary assumptions are introduced to flesh out the model. And not only does it justify my program, but it also sheds new light upon why deficit spending on a very large scale will sometimes be socially optimal, for reasons far transcending Keynes’ concept of boosting aggregate demand during recessions.

There are two steps in implementing the logic required of the Philosopher King. Just as the captain of the rocket ship must identify the least-time trajectory and determine the right setting of the control instruments (e.g., the thrusters) at each point in time, to remain on course, so must the Philosopher King.

Step 1: Determine the right trajectory of economic outcomes that are optimal (the satisfaction-maximizing time paths of consumption and of private and public sector investment).

Step 2: Identify the right setting of the government control knobs (choice of public investments, of tax rates, and of fiscal deficit levels) that is required to maximize public satisfaction. As in the example of the rocket ship, there will be a unique set of values for the control knobs at each point of time that will maximize public satisfaction.

President: That is quite a mouthful. Is there a graphic way of portraying all this?

Economist: Yes. Figures 2.4(a) and 2.4(b) should make all this quite clear. Please understand that any data appearing here are hypothetical. The analysis is purely illustrative.

Figure 2.4 (a) Arrow-Kurz Unification of Macroeconomics (b) Optimal Tax Rates

Source: Strategic Economic Decisions, Inc.

To maximize welfare first requires solving simultaneously for the three optimal trajectories-over-time seen in Figure 2.4(a), as well as for the optimal tax rates shown in Figure 2.4(b). Arrow and Kurz showed why this is necessary, and how to do so. The first graph represents the optimal GDP-share of total investment spending (private and public). In solving for this, you also solve for the optimal share of consumption and savings. But this is not shown.

President: Does this first trajectory offer a way of showing that a nation may be saving and investing too much of GDP (like China), or too little of GDP (like the United States)—familiar criticisms of both nations?

Economist: Yes. There is an optimal in-between level of investment, one for each country. Traditional growth theory focused on determining this path. But prior to the Arrow-Kurz theory, no distinction was made between private sector investment and public sector investment. But such a distinction must be made: Both are crucial for public welfare (the right number of houses and the right amount of public transport). Yet the logic for determining the optimal mix of both is highly complex since the meaning of “high return private investment” is so different from “a high-return public investment.”

President: Isn’t that what the second trajectory in the middle graph is all about?

Economist: Yes. Here you must solve for the optimal share of private versus public investment, given the optimal GDP-share of total investment. Arrow and Kurz provided the logic to do so. This optimal share will differ across time as shown, since during some eras, returns are greatest to private sector investing, but during other eras, returns are greatest to public sector investing.

President: Aren’t the two problems indicated in the first two graphs interdependent with each other? That is, wouldn’t the determination of the optimal total share of investment spending (as opposed to consumption) depend on the mix of returns available in both the private and public arenas?

Economist: Yes, indeed. This is why you must solve for both trajectories simultaneously. Arrow and Kurz demonstrated how to do so. Note that you are not solving here for two optimal numbers, but rather for two optimal trajectories over time. Moreover, these trajectories are not “fixed,” but rather are contingent upon how external “state” variables evolve over time. This complicates matters still further.10

President: Well, all this does seem to make sense, given how you explain it. There will be times when private investment seems most “needed,” for example, young GIs returning from World War II all needing houses in which to rear their baby-boomer children. And there are times like today when there seems to be a surfeit of houses, and a dearth of good public transportation to move people around.

Economist: This is exactly the point, and you have given a good example that I will come back to.

President: And you will argue that we are right now at a juncture where higher returns (properly measured!) will accrue from massive public sector investment rather than from private sector investment, correct?

Economist: Yes, indeed, and this is true because of our deferred infrastructure maintenance over the past half-century, and because of our overinvestment in housing. Regrettably, the nation has acted in a way that fulfilled John Kenneth Galbraith’s prediction of a nature showcasing “private splendor and public squalor.”

Rethinking the Role of Deficit Spending from Scratch

President: How does a decision to run large government deficits fit into all this? The third and last graph of Figure 2.4(a) seems to focus on deficits.

Economist: Yes. This graph shows the optimal size of the government deficit over time as a share of GDP. The reason that fiscal deficits are part of the analysis is that public borrowing is sometimes required if social welfare is to be maximized. What is not shown (for the sake of simplicity) is that there is an analogous trajectory of private sector borrowing indicating the times when the private sector should run deficits versus surpluses. Moreover, the two different deficits/surpluses are interrelated at a deeper level. There will be times when the private sector should run a surplus that can fund a public sector deficit, and vice versa. Remember, the money to fund any deficit must come from somewhere as we noted in the context of equation (2.3). But let’s avoid this complication here.

President: What does all this say to people who instinctively disapprove of government deficits?

Economist: It says: “Discard your prejudices against government deficits. To begin with, if you are philosophically prejudiced against public sector borrowing, no matter how high the rate of return might be, than you will logically be compelled to reject private sector deficits as well, since the roles of the two deficits are symmetrical in maximizing welfare, when properly understood.” Equivalently, you could say: “To claim that government deficits are intrinsically bad is tantamount to telling families in the private sector that they should never take out a mortgage, since debt is bad. If they do not have enough saved up to buy a house entirely with cash when the children are growing up, then they should lump it and rent.” A final way of expressing this is: “If you are the Philosopher King and you are attempting to maximize societal welfare, then you can do so only if the private and the public sectors borrow and save optimally over time. Their situations are symmetrical.”

To be sure, excessive borrowing can be bad, and should be limited. This is achieved within the Arrow-Kurz framework by imposing constraints on the accumulation of debt over time (whether household or government debt). Thus, when the optimization required by the Arrow-Kurz logic is carried out, while there may be short periods when huge deficits are required, the overall accumulation of debt over time will not be allowed to exceed appropriate levels in either the private or public sector. Such constraints are needed because, as we are reminded over and over, excessive borrowing either by government or by households amounts to an externality, or public bad. You will read more on this in Chapter 4 when I examine the lessons of the global financial crisis of 2007–2009. I demonstrate that excessive leverage was the primary culprit, not “greed” or regulatory incompetence.

President: So all in all, what Professors Arrow and Kurz showed is that it is both necessary and possible to determine all three of the optimal paths sketched in your graph: the optimal level of investment spending over time, the optimal mix of private and public investment over time, and the optimal level of the fiscal deficit or surplus over time. And all three must be determined simultaneously because all three are interdependent. And everything here follows from the appealing principle of “maximizing the greatest good for the greatest number,” so to speak. Is that it?

Economist: Yes, and that principle could also be dubbed “maximize efficiency” or “minimize waste.” They are all one and the same.

President: But what’s in your Figure 2.4(b)? How do taxes fit in here?

Economist: Oh, I forgot to mention this. It turns out that Arrow and Kurz accomplished even more. Remember, there are two steps involved in our overall problem, as I mentioned up front. The first is to determine the rocket’s optimal trajectory to the moon and back, the three trajectories of Figure 2.4(a) in the case of economics. The second task is to determine the precise settings of the control instruments (the “thrusters”) that make it possible to adhere to the optimal trajectories at each point along the way. The authors showed how to determine within economics the right setting of the control instruments needed to achieve this goal, that is, the right level of total taxation and the right mix of the different tax revenues available. This is indicated in Figure 2.4(b). Once again, these optimal tax rates must be determined simultaneously with the three trajectories of Figure 2.4(a).

The optimal total tax and the optimal composition of taxes will be dynamic, of course. These variables will be trajectories that change along the journey, and must be monitored and altered en route. Breaking further new ground, the authors determined the exact conditions under which a dynamical economic system can be optimally controlled, that is, if and when a suitable setting of the various tax rates exists.

The issue here is known as “macro-controllability,” a branch of economics separate from public finance, fiscal policy, and growth theory. Arrow and Kurz extended the theory of controllability to incorporate these other three fields within it, and in doing so achieved perhaps the greatest synthesis or “unification” within macroeconomic theory that has ever been achieved. Yet because their book was necessarily mathematical, it remains largely unknown outside the hallowed halls of academia.

President: All this is mind-boggling to a noneconomist like myself, and is completely new. It gives new meaning to the concept that “everything depends upon everything else.” You stated earlier that their work paves the way for a much deeper understanding of when and why very large deficits can sometimes be needed. Is there a deeper message lurking here about the appropriateness of deficit spending? After all, you are proposing to keep the U.S. fiscal deficit at today’s record high level of 10 percent of GDP, at least as conventionally measured. Are Arrow and Kurz on the same track as Lord Keynes was in proposing deficit spending as a remedy for recessions? It is odd that you have not mentioned Keynes in all this.

When Very Large Deficits Are Justified: A Unification of the Arrow-Kurz and Keynesian Theories

Economist: There is a very good reason why Keynesian theory has not been mentioned. The Arrow-Kurz theory was set forth within a “classical” economic framework of full employment and no business cycles. Thus there is no need for anything Keynesian within it. The reason why there are deficit cycles that look Keynesian (as shown in the third graph) is because the optimal mix of private versus public sector investment returns fluctuates across time. This mix will sometimes call for greater investment spending (and hence borrowing up front) in the public sector and less in the private sector, and at other times it will call for the reverse. For example, there will be times when the private sector should borrow a lot for the mortgages of those returning home from war to build houses for their young families. There will be other times when higher returns are reaped by investing in deteriorating public infrastructure as opposed to more McMansions. This important “cycle of investment opportunities” can perfectly well coexist with a full employment economy free of business cycles.

President: So running deficits to pump up demand in business cycle downturns was not part of the Arrow-Kurz fiscal story at all?

Economist: No, it was not. But because of much more recent work by Kurz and others that extend “classical” economics to include fluctuating animal spirits and business cycles, their original theory can in principle be extended to include Keynesian concerns. In this case, the optimal government deficit sketched in the third graph in Figure 2.4(a) will be large when animal spirits and hence private investment are low in a recession. Conversely, the fiscal deficit will be small or even negative when animal spirits and private investment are high. That would be purely Keynesian.

President: So can the two different rationales of Keynes and of Arrow-Kurz for running government deficits be combined?

Economist: Yes, and this is the main and final point to be raised. If someone asks: “Under what sets of circumstances will the public welfare be maximized by running very large fiscal deficits?” The answer is: “To maximize social welfare, the fiscal deficit should be very large if the returns to public investment are much higher than in private investment, and also if animal spirits in the private sector are flagging so that Keynesian cyclical stimulus is needed to increase aggregate demand.” And vice versa for when there should be a significant fiscal surplus. All these relationships are symmetrical in that they work in reverse.

President: So you claim that this pair of circumstances justifying large deficits accurately describes the situation in the United States at the moment: The potential returns from infrastructure spending are large because of past neglect and because of new opportunities, and because animal spirits are poor throughout the private sector due to the global financial crisis, to 30 years of excessive binge-borrowing by households, to growing retirement worries, and to soaring medical costs.

Economist: Correct, but do not forget we are talking about large “good” deficits that arise from productive spending—not from bad spending that further burdens our children with more debt. Remember, one of the four main objectives in preventing a Lost Decade must be to calm the bond markets. At present, this is not being done. And this is something that neither Keynes nor Arrow and Kurz discussed. It is a new constraint.

For any economists who advise you and who might be interested, the argument I just made as to when very large deficits are needed is sketched more formally in the Appendix that follows. I included this since my arguments here will be unfamiliar to most economists.

Thank you, Mr. President, for listening to my proposal. I hope I have fulfilled my principal goal: to demonstrate the existence of win-win policies to prevent a Lost Decade, policies that are consistent with all four desiderata laid out at the start, and that are justified more fundamentally by the single axiom of “maximizing human welfare.” Ultimately, my message is one of optimism: There exists a policy solution sufficiently win-win in nature to break partisan gridlock on the entire deficit spending issue that has now has paralyzed debate in the country. It really is possible for you to have your cake and eat it too, Mr. President.

Appendix: The Case for Very Large Deficits and/or Surpluses

Consider an economy with public goods and externalities of the A-K (Arrow-Kurz) variety. Suppose private consumers and investors have the objective of selecting state-dependent policies that maximize their expected utility. And suppose their government watchdogs select policies that maximize societal felicity, that is, the sum of utilities across people.

Next let us introduce two cycles into this optimization model. The first cycle is a Markov process consisting of two dominating states, A and B. Let state A denote a world in which all consumers possess a McMansion, yet public transportation is so dilapidated that citizens living in these dream houses cannot get to work without an undue waste of time and energy. In state B, the situation is reversed and recalls Japan in 1970: There were bullet trains that were the wonder of the world, but the Japanese lived in small houses lacking basic amenities. Then the solution to the Arrow-Kurz optimization problem will be to shift resources from the private to the public sector when the system is in state A where transportation is needed, and private housing is not. And this transfer will be reversed when the system is in state B where private housing is needed, and public transportation is not.

Now let us add a simple two-state Keynesian business cycle to the Markov model. There are now two new states C and D. State C denotes flagging animal spirits and recession (e.g., slack private demand), whereas state D represents robust animal spirits and a private-sector boom. Recall that there are no business cycles in the neoclassical A-K model. We can then construct an nth order Markov process consisting jointly of the investment-cycle state and the animal-spirits state. In this model, the system will migrate at any point in time to one of the four possible joint states: AC, AD, BC, or BD.

When this A-K model is suitably constrained and optimized over the relevant space of policy functions, utilizing a generalization of the Pontryagin maximum principle from optimal control theory, then social welfare will be maximized by running very large public sector deficits when joint state AC is visited (high returns from public transport investment and a low level of animal spirits in the private sector), whereas a very large fiscal surplus will be optimal when joint state BD is at hand (high returns from private-sector housing investment and a high level of animal spirits).

This model provides completely new support to the concept that large fiscal deficits are indeed optimal in certain environments, and are required for investment-cycle reasons far transcending Keynesian animal-spirits reasons. Importantly, there is a duality whereby, in some states of the system, the household will be a net borrower (as when everyone after a war needs a mortgage to buy a house in which to raise a family), and in other states the household will be a net lender. But the same is true of the public sector, which itself will run deficits in states where it is utility-maximizing to do so, and surpluses in other states of the system. There seems to be little or no awareness of this private/public sector symmetry.

In this regard, the great achievement of the A-K contribution in my eyes was to dissolve the putative difference between “macro” (fiscal) and “micro” economics, and to show how there is one and only one fundamental problem in all of economics: to identify and implement an optimal allocation of all resources (private and public) across time so as to maximize public well-being.

1. By utilizing formal statistical analysis, you can learn that the decade-by-decade volatility of GDP declined by about 75 percent over the entire twentieth century (some of the earlier decades are not shown in the figure). There are five interesting reasons why this occurred. First, having been invented in the 1930s and 1940s, macroeconomic policy became utilized to stabilize business cycles from the late 1950s on. Second, the inventory cycle was tamed due to the advent of “just-in-time” inventory management techniques. Third, the manufacturing sector is inherently more volatile than the service sector, and the latter grew at the expense of the former, further stabilizing the economy. This is particularly true in the case of labor utilization, where an accelerating share of the workforce worked in services between 1945 and 2000. Fourth, family income was stabilized by the advent of two-income families. Fifth and last, both the level of household wealth and the access to credit increased. This provided a cushion that helped families to get through difficult times more easily so that consumption became less volatile.

2. Equation 2.1 is primarily useful for forecasting one- or two-year changes in GDP, and not much longer. This is because the terms in it refer to the sources of demand, not supply. More specifically, this formalism abstracts from changes in the productivity of labor, in the growth of the workforce, in total factor productivity, and in other longer-run supply-side factors that impact GDP growth along with demand. This point was stressed to me by Professor Benjamin Friedman at Harvard.

3. We are ignoring the issue of “statistical inconsistencies,” such as unreported income, which often causes NI to be less than NP ex post. However, such discrepancies have little bearing on ex ante forecasts of either variable, which is why we shall ignore them.

4. The reader looking for a formal definition of National Income can now see that, by breaking out the Net Private Savings term in equation (2.3) into NI – Private Spending, and substituting this into (2.3), and then by rearranging terms within (2.3) so that NI ends up alone on the left-hand side, we arrive at an equation defining NI itself. The result here is of no importance in this discussion. But this proves that (2.3) is in effect the equation of NI. This is disguised in the way I have written the identity.

5. For example, suppose that the U.S. trade deficit does not change over time. Then the dollar value of this net inflow will not change regardless of changing foreign asset preferences. For it must continue to equal and indeed to finance the unchanged trade deficit. Given this flow-of-funds constraint, known as the Capital Account = Current Account identity, it turns out that, when foreigners become disenchanted with buying more U.S. assets, the value of the dollar drops, whereas U.S. interest rates do not generally rise. At the new lower exchange rate, foreigners are spending less of their money buying U.S. assets, but they are spending the same amount in U.S. dollars. Thus, there is no reduction in capital inflows in dollars so that U.S. interest rates are not driven up, as is supposed will happen by virtually everyone. Rather, the value of the currency gets driven down to a point where reduced currency risk rather than higher yields is what attracts foreign investors. The currency does virtually all of the work. This is known as the interest-rate/exchange-rate adjustment process, and it is true in theory and in practice. It is highly counterintuitive, but very useful in real-world forecasting. A lower exchange rate as opposed to a higher interest rate will impact GDP growth in opposite ways, the former boosting growth, the latter depressing it.

6. See Bold Endeavors: How Our Government Built America, and Why It Must Rebuild Now, by Felix G. Rohatyn (New York: Simon & Schuster, 2009).

7. In what follows, I shall ignore monetary policy and focus on fiscal policy (taxes and fiscal deficits). For it is fiscal policy that is central to my proposed policies. Arrow and Kurz also only focus on fiscal policy, as the title of their book makes clear.

8. The model is that of optimal control theory, a discipline largely developed in the 1940s and early 1950s. There are two versions of the model: One is based on the Pontryagin maximum principle in optimal control theory, and the other on the principle of optimality in dynamic programming theory. Chapter 2 of the Arrow-Kurz book reviews both in mathematical depth. The former theory is for continuous time models, whereas the latter is for discrete time models.

9. In formal economics-speak, the objective function is the felicity functional justified at length in the Arrow-Kurz book. This is an interpersonal extension to society of the intrapersonal mandate that each person on his own will attempt to maximize his utility for what is at stake.

10. More formally, each trajectory is a so-called policy function indicating what decision is best at each “state” of the system as it evolves over time.