In this section, we discuss trends in transportation investment in the following sequence: overall transportation investment, transportation investment by sector, transportation investment by asset type (i.e., infrastructure and rolling stock) and by sector of investors (i.e., government, business, and household sector), business investment by category (i.e., transportation and non-transportation investment), and business investment in transportation by industry (i.e., transportation vs. non-transportation industries).
The overall transportation investment accounted for more than 6% of GDP for most years during the period of 1977 to 2000 (Figure 1). Of this overall transportation investment, investment (including household expenditure) in rolling stock accounted for 83.3%, and that in transportation infrastructure and in other transportation equipment, 14.2% and 2.5%, respectively. As such, the trend of investment in rolling stock dominates the trend in transportation investment. It is also noteworthy that, whereas infrastructure investment appears to have been steady at around 0.9% of GDP since 1981 when it dropped below 1% of GDP, the overall expenditure in rolling stock (net of government purchase for defense) fluctuated along with the economic cycle. For example, the year-by-year drops in overall expenditure in rolling stock during the period of 1979 to 1982 echoed the repeated business downturns during that period (Figure 2). The picture for 1991 tells a similar story, whereas that for 1996 reflects a brief economic stagnation in 1995. Finally, business investment in transportation equipment (other than rolling stock) as a share of GDP, although insignificant, appeared to increase steadily for the second half of the 1990s, which may indicate technological advancement within the transportation industry during that period and warrants further study (Figure 1).
Figure 3 illustrates transportation investment by sector, namely, households, private business, and government. As a share of GDP, government investment, which consists of infrastructure and rolling stock, stayed almost the same, averaging 0.9% for the entire period, while private business investment, which includes infrastructure, rolling stock and other equipment, declined until 1990 and showed a consistent increase since 1991. Transportation investment of the household sector has been consistently greater than that of the public and private business sectors combined during the period under consideration. As a share of GDP, transportation investment by the household sector averaged 3.7%, fluctuating between a low of 3.1% and a high of 4.5% during 1977-2000.
The role of government, private business, and households in transportation investment vary by asset type and mode of transportation (Figures 3 and 4). The public sector provides most transportation infrastructure with the exception of railroads and pipelines. For example, of the total government transportation investment, infrastructure accounted for an average of 89% for the period 1977-2000. The private business sector invests heavily in rolling stock and operates on publicly provided infrastructure. Private business investment in rolling stock and other equipment used for transportation activities accounted for about 92% of the total private sector transportation investment. The remaining 8% went to transportation infrastructure. A large proportion of the business infrastructure investment has been used for railroads and pipelines, which accounted for about 60% on average during 1977-2000. The household sector doesn't invest in transportation infrastructure and hence its investment is entirely in rolling stocks (i.e., motor vehicles, trucks and vans, airplanes, boats, etc.).
Figure 4 provides an enlarged picture of investment in transportation infrastructure in terms of time trend and by sector. Figures 5-7 further illustrates this type of investment by transportation mode and by sector of investors. Four observations can be drawn from these figures.
First, government is the predominant investor in transport infrastructure (Figure 4). Therefore, government investment clearly dominates the trend of investment in transportation infrastructure with only a few exceptions (e.g., in 1978, 1986, and 1987). For the period of 1977 to 2000, the government share in infrastructure investment, on average, is 97% for highways, 94% for transit, 81% for airports, and 84% for water transportation (Figure 5). While the business sector accounts for almost 100% of infrastructure investment for railroads and pipelines, total infrastructure investment for these two modes accounts for less than 0.1% of GDP since 1986 (Figure 5). With its minor share, the business sector's investment in transportation infrastructure as a proportion of GDP had been dropping since 1977 when it accounted for 0.21% of GDP and stabilized since 1987 around 0.1% of GDP, with a brief rise to 0.14% of GDP for 1998 and 1999 (Figure 4).
Second, investment in transportation infrastructure lagged behind the business cycle. Overall investment in transportation infrastructure peaked in 1980, followed by a sharp downturn that bottomed out in 1984 and has fluctuated ever since, with a strong pickup in the late 1990s when both government and business sectors increased their investment (Figure 4). It is interesting to note that overall investment in transportation infrastructure lagged behind the business cycles. For example, whereas overall transportation investment, of which about 80% is in rolling stock, started to drop in 1979 when business contraction was well underway (Figure 2), investment in transportation infrastructure rose in 1979 and receded in 1980 and 1981 when the economy showed a brief revival (Figure 4). This implies that infrastructure investment, particularly that funded by the government sector, is an induced variable of past economic performance; it could be also a result of government budget decisions, which is often counter cyclical.
Third, highway investment, on average, accounts for over 65% of total investment in transportation infrastructure for the period of 1977 to 2000 (Figure 5 and 6). Consequently, the trend in highway investment dominates the observed trend in the overall infrastructure investment.
Finally, despite all the fluctuation, infrastructure investment in non-highway transportation modes changed significantly in relative shares. As illustrated by Figure 7, as a percentage of GDP, investment in air transportation (mainly airports) in 2000 (0.11%) was almost double that in 1977 (0.056%). Similarly, though less significantly, investment in both water transportation and public transit as a percentage of GDP increased by more than 20% from 1977 to 2000. In contrast, investment in railroad as a percentage of GDP has decreased over time by more than 60% and that in pipelines dropped by 90%.
Figures 8-11 illustrate several characteristics of investment in rolling stock measured as a percentage of GDP.
First, households are the predominant investors in rolling stock, particularly motor vehicles (Figure 9). This is well documented in the National Account (for example, see the BEA website, Table 8.8U for Auto Output, at http://www.bea.gov). The annual change in household-dominated investment in rolling stock appeared to be closely related to the change in the annual growth rate in GDP (Figure 10). This is consistent with the close association between household spending and overall economic performance. In other words, as the growth in GDP speeds up, so does household purchase of rolling stock as a proportion of GDP (sometimes with a short time lag), and vice versa.
Second, compared with their investment in transportation infrastructure, government and business investment in rolling stock is more cyclical, although with more evident time lag than household purchase of rolling stock. On the other hand, similar to their investment in infrastructure, government and business investment in rolling stock showed a steady upward trend since 1992. This trend requires a further investigation of its economic driving forces and consequences.
Finally, the majority of investment in rolling stock is in motor vehicles, including automobiles, trucks and buses. As a result of this, investment in motor vehicles (for highway/transit modes) dominates the trend of investment in rolling stocks (Figure 11). Investment in motor vehicles, or rolling stock for highway and transit use, has been well above 90% of the total gross investment in rolling stock since 1983.
Figure 8 provides a summary of investment in rolling stock by sector and by transportation mode. As described above, the figure shows that household purchase of automobiles (for the highway mode) on average accounts for about 70% (=3.71%/5.24%) of gross investment in rolling stock. This is consistent with the study on household production of transportation services (Chen, Fang, Han and Sloboda, 2002). Since household purchase of motor vehicles is one of the major components of gross personal consumption expenditure (PCE) and follows closely the trend in GDP growth, it contributes significantly to the fact that PCE is used as a critical component of many economic indicators (e.g., manufacturers' new orders for consumer goods and materials, the personal consumption expenditure deflator for money supply, and consumer installment credit to income ratio, etc.).
The term "business investment" in our report is equivalent to the term "gross private fixed investment" used in the input-output (I-O) accounts. For our purposes, business investment, or gross private fixed investment (GPFI), can be classified as transportation and non-transportation investment. As mentioned before, business investment in transportation includes all the investments made by the transportation industries and investments in rolling stock made by non-transportation industries; non-transportation investment includes the rest. Compared with non-transportation investment, transportation investment is more sensitive to the business cycle as measured by growth in GDP. Figure 12 illustrates this comparison. In other words, compared with non-transportation investment, transportation investment is more suppressed by economic contraction and more buoyed by economic recovery and boom. A possible explanation is that, compared with non-transportation investment, transportation investment (largely rolling stock as mentioned above) is more flexible in terms of its utilization and useful life, hence replacement cycle. On the other hand, several notable overshoots in transportation investment, particularly those appearing to go against the business cycle (e.g., in 1990-91), require further investigation.
Transportation investment made by non-transportation industries accounts for 6.7% of GPFI and 1.1% of GDP, respectively, whereas that by the transportation industry accounts for 4.2% and 0.7%, respectively (Figure 13). In other words, transportation investment made by non-transportation industries (all in rolling stock) is about 60% more than that made by transportation industries (including infrastructure, rolling stock and other equipment). It is interesting to look at these statistics in reference to the transportation satellite accounts (TSA). As indicated in the 1992 TSA (BTS, 1999), in-house transportation services provided by non-transportation industries were about 60% of the for-hire transportation services provided by transportation industries (i.e., in-house accounted for 1.9% of GDP; for-hire, 3.1%). This rough comparison indicates that the intensity of use of transportation capital (largely rolling stock) is higher for the transportation industries than that for non-transportation industries.
In summary, for the period of 1977 to 2000, overall transportation investment (including household purchase of rolling stock) on average accounted for more than 6% of GDP. Of this overall transportation investment, investment in rolling stock accounted for 83.3%, and that in transportation infrastructure and in other transportation equipment, 14.2% and 2.5%, respectively. Second, while households are the primary purchaser of rolling stock, government is the dominant investor in transportation infrastructure, except for railroads and pipelines, in which the business sector appears to be the exclusive investor. Third, besides the dominant share of highways in total infrastructure investment, which averaged 66%, the relative share of infrastructure investment among other modes changed significantly during 1977-2000, with air investment almost doubling and pipelines investment dropping by about 90%. Fourth, whereas the overall transportation investment (of which 83% is rolling stock) closely echoed the business cycle, investment in transportation infrastructure evidently lagged behind the business cycle. Fifth, within business investment, transportation investment is more sensitive to the business cycle compared with non-transportation investment. Finally, transportation investment made by non-transportation industries is greater than that by transportation industries. However, the intensity of use of transportation capital, particularly rolling stock, is much higher in transportation industries than in non-transportation industries. Readers should note that these findings require further study of their economic and policy background to facilitate future budgeting for transportation investment.
It is known that government invests in railroad infrastructure. Most of the public investment is, however, allocated to commuter rail and subways, which are counted in transit mode.