This article has multiple issues. Please help improve it or discuss these issues on the talk page. (Learn how and when to remove these template messages)(Learn how and when to remove this template message)
Investment-specific technological progress refers to progress that requires investment in new equipment and structures embodying the latest technology in order to realize its benefits.
To model how something is produced, think of a box that in one end takes in inputs such as labor (employees) and capital (equipment, buildings, etc.) and in another end spits out the final good. With this picture in mind now one can ask, how does technological progress affect production? One way of thinking is that technological progress affects specific inputs (arrows going in) such as equipment and buildings. To realize the benefits of such technological change for production, these inputs must be purchased. So for example, the advent of the microchip (an important technological improvement in computers) will affect the production of Ford cars only if Ford Motor Co.'s assembly plants (the red box) invest in computers with microchips (instead of computers with punched cards) and use them (they are one of the arrows going in the box) in the production of Mustangs (the arrow coming out). As the name suggests, this is investment-specific technological progress---it requires investing in new machines or buildings which contain or embody the latest technology. Notice that the term investment can be general: not only must a firm buy the new technology to reap its benefits, but it also must invest in training its workers and managers to be able to use this new technology (Greenwood & Jovanovic 2001) .
Identifying investment-specific technological progress is important, because knowing what type of technological progress is operating in an economy will determine how someone (should) want his or her tax dollars to be spent and how he or she may want to invest his or her savings (Gort et al. 1999). If "investment-specific" technological change is the main source of progress, then one would want his or her dollars spent on helping firms buy new equipment and renovate their plants, because these investments will improve production and hence what you consume. Furthermore, one may want to help pay for current employee training in using new technologies (to keep them up to date) or subsidize the education of new employees (who will enter the job market knowing how to use the new technology). So, the type of technological progress will also matter for unemployment and education issues. Finally, if technological progress is "investment-specific" you may want to direct your money towards the research and development (R & D) of new technologies (like quantum computers or alternative energy sources) (Krusell 1998).
More generally, why is any type of technological progress important? Technological change has made humans lives' easier. Because of technological progress, people can work less, make more money and enjoy more leisure time (Greenwood & Vandenbroucke 2006). Women have been able to break away from the traditional "housewife" role, join the labor-force in greater numbers (Greenwood et al. 2005) and become less economically dependent on men (Greenwood & Guner 2009). Finally, technological progress has been shown to affect the fall in child labor starting around 1900 (Greenwood & Seshadri 2005). Figure 1 illustrates this last point: in 1900 child labor's share of the paid labor force began to fall.
An example of investment-specific technological progress is the microwave oven. The idea of the microwave came to be by accident: in 1946 an engineer noticed that a candy bar in his pocket had melted while working on something completely unrelated to cooking (Gallawa 2005). The development of this good, from melting the candy bar to the home appliance known today, took time and the investment of resources to make a microwave small and cheap. The first microwave oven cost between 2000 and 3000 dollars and was housed in refrigerator-sized cabinets (Gallawa 2005)! Today, almost any college student can enjoy a 3-minute microwaveable meal in the smallest dorm room. But a microwave's uses do not stop at the dorm room. Many industries have found microwave heating advantageous: it has been used to dry cork, ceramics, paper, leather, and so on (Gallawa 2005). However, for either college students or firms to reap the benefits of quick warming, they must first "invest" in a microwave oven (that "embodies" the technological advance). To realize the benefits of investment-specific technological progress you must first invest in a technology that embodies it.
While measuring technological progress is not easy, economists have found indirect ways of estimating it. If "'investment-specific'" technological progress makes producing goods easier, then the price of the goods affected (relative to the price of other goods) should decrease. In particular, "investment-specific" technological advance has affected the prices of two inputs into the production process: equipment and structures. Think of equipment as machines (like computers) and structures as buildings. If there is technological progress in the production (or creation) of these goods, then one would expect the price of them to fall or the value of them to rise relative to older versions of the same good.
Figure 2 (the pink line) shows how the price of new producer durables (such as equipment) in the United States relative to the price of new consumer nondurables (like clothing) has consistently declined over the past fifty years (Gort et al. 1999). To calculate the relative price of producer durables divide the price that firms pay (for the durable inputs of production) by the price that a regular consumer pays (for things like jeans). People use relative prices so they can say how many units of equipment can be bought instead (or in terms) of buying one unit of consumer goods. Figure 3 (the pink line) says that over time, firms have been able to buy more and more units of equipment instead of one unit of consumption, especially when taken into account that the quality of equipment being acquired has increased (a computer today is much faster than a computer five years ago and that should be taken into account when comparing their prices). When changes in quality are not taken into account (which is wrong) it looks like the price of equipment has not decreased as much (see the black line in Figure 2).
Measuring the price of structures is more complicated than measuring the price of equipment, but economists have again been able to get an idea of how much progress there has been in structures (such as buildings). One approach is that if newer buildings were constructed or designed using newer technologies then they should be worth more than older buildings (because they embody the new technology (Gort et al. 1999). In particular, they should rent for more. As Figure 3 shows, this is true. Renting a square foot in a new building is much more expensive than renting a square foot in a building forty years old. So it must be the case that you are paying for a nicer, more functional and maybe even safer building.
Figures 2 and 3 suggest that investment-specific technological change is operating in the US. The annual rate of technological progress in equipment and structures has been estimated to be about 3.2% and 1%, respectively (Gort et al. 1999) (Greenwood et al. 1997).
In the second section it was mentioned that "investment-specific" technological change is important since it will affect production (both in quality and size). An important question then is, just how much "bang for your buck" do you get with "investment-specific" technological change? The answer is quite astounding; economists have found that 37% of growth in United States output (production) is due to technological progress in equipment and 15% is due to technological progress in structures (Gort et al. 1999) (Greenwood et al. 1997). All in all, more than half (37% + 15% = 52%) of the growth of the United States economy is due to "investment-specific" technological change (Gort et al. 1999) (Greenwood et al. 1997).