The Emerging Geography of Electric Vehicle Production in North America: Revolution or Evolution?

Introduction and summary

The automotive industry has embarked on a major transition from manufacturing gas-powered vehicles to producing
electric vehicles (EVs).¹ In this article, we parse out the
implications of this shift for the industry’s
production
footprint across North America, comprising Canada, Mexico, and the United States, between 2023 and 2029. We
compare
the emerging geography of battery electric vehicle (BEV) production facilities with the existing distribution of
internal combustion engine (ICE) vehicle production facilities across the region. Two other types of
vehicles—hybrid
electric vehicles (HEVs) and plug-in hybrid electric vehicles (PHEVs)—are also included in our analysis here
because
they are forecasted to account for significant shares of production in the years ahead.

With this article, we build on our 2022 Economic Perspectives article (Klier and Rubenstein, 2022a),
which was
devoted primarily to background materials related to electric vehicle development. Our earlier article explained
that agglomeration economics and economies of scale are shaping the geography of BEV production; these two key
factors also underlie the location decisions for ICE vehicle production. It included a detailed description of
the
differences among the various propulsion systems; a brief history of the regulatory environment that is
encouraging
the development of EVs; and the principal obstacles to consumer adoption of BEVs. Here we complement our
previous
analysis by drawing on extensive forecasts for the production of light vehicles (that is, cars and light
trucks), as
well as their propulsion systems (gas engines, batteries, or a combination of them). These forecasts allow us to
discuss changes in the industry footprint at a much richer level of detail than before, including our being able
to
distinguish between light vehicle production by all major propulsion systems.²

In this article, we are concerned with possible changes to the footprint of the automobile industry in North
America
over the medium term—that is, over the next five to seven years. To guide our analysis, we rely on comprehensive
production forecasts by S&P Global Mobility, a well-respected provider of such data.³ S&P Global Mobility
utilizes
deep industry expertise to allocate industry production of light vehicles to specific assembly plants, as well
as to
forecast the sourcing of key vehicle systems, such as engines and batteries. We utilize economic geography
concepts
to analyze S&P Global Mobility’s detailed production forecasts for individual factories in the United States,
Canada, and Mexico (all three countries that make up North America produce light vehicles). The figures provided
in
this article are derived from a careful parsing of current and forecast data provided by S&P Global Mobility.

The time horizon of the S&P Global Mobility forecasts allows us to look ahead along an expected path for the auto
industry’s transition toward electrification over the medium term. Naturally, there is some uncertainty around
the
pace of this transition for the industry (it’s difficult to estimate when the vast majority of new vehicles will
be
EVs). Some of this uncertainty stems from the fact that incumbent light vehicle producers, on the one hand, and
start-up pure-EV producers, on the other, are facing the transition differently. Incumbent producers need to
accommodate a switch from one type of product to a different type of product in the context of mass production;
during the transition toward electrification, funding for the emerging side of the business (EV production) will
likely be generated by the declining side of the business (ICE vehicle production). The new entrants to the auto
sector, however, face a different challenge: These pure-EV manufacturers must cover the high start-up costs in
this
capital-intensive industry and survive until profitability is reached.

In the next section of this article, we briefly review the current geography of light vehicle production in North
America. More than 90% of light vehicles produced in the region in 2023 included a gasoline engine (see figure
3).
In 2023, most vehicle assembly plants producing those vehicles (namely, ICE vehicles, HEVs, and PHEVs) and the
accompanying engine plants within North America were located in a corridor between southwestern Ontario, Canada,
and
the Gulf of Mexico, known as “auto alley,” plus in a second cluster in central Mexico. As part of this analysis,
we
map out and describe the locations of assembly plants for the specific types of vehicles (with different
propulsion
systems)—that is, BEVs, PHEVs, HEVs, and ICE vehicles—as of 2023.

After that, we describe the future geography of light vehicle production across North America, based on forecast
data
from S&P Global Mobility for 2029. We analyze how the anticipated shifts in product mix between 2023 and 2029
are
expected to affect the locations of light vehicle assembly plants over this span. It turns out not very much, at
least when considering the production footprint across North America from a broad regional perspective (which we
elaborate on later).

Next, we compare the emerging distribution of battery plants with the distribution of engine plants.⁴ Since the
publication of our 2022 Economic Perspectives article, we have obtained much more detailed
information
on the
location decisions made by the manufacturers of EV battery components, such as battery cells and battery packs.

Finally, we conclude the article with a discussion about the emerging footprint of electric vehicle production
across
North America. We find that despite the massive changes expected in the mix of vehicle types produced, the
emerging
spatial pattern of BEV assembly and battery production is strikingly similar to the existing pattern for ICE
vehicle
assembly and engine production across the broader region. That said, we recognize that new battery plants have
been
(and likely will be) constructed in different communities than those that are home to existing engine plants.
Thus,
at the local scale of individual cities and towns, the opening of battery plants—as well as the possible closure
of
engine plants or reduced production activity at them—can portend changes in local economic fortunes.

We ended our 2022 Economic Perspectives article with a tentative suggestion that the footprint of
BEV
production
across North America might not be much different from the prevailing footprint of ICE vehicle production in the
region. In this article, we reinforce that conclusion, using data that are much richer in detail than those we’d
used for our 2022 article. Indeed, our findings based on the S&P Global Mobility data in this article are more
refined and nuanced than those discussed in that previous work. For example, it is important to note that the
two
“intermediate” technologies that power HEVs and PHEVs are playing important roles in the emerging geography of
light
vehicle production—at least through the end of the current decade—in part because both of these types of
vehicles
still require an internal combustion engine.

Current footprint of light vehicle assembly plants

Most assembly plants are located in an area within North America known as auto alley—a corridor chiefly within
the
United States between the Great Lakes and the Gulf of Mexico, with an extension at its northern end into
southwestern Ontario, Canada (Klier and Rubenstein, 2022a). It became the principal location for auto production
beginning in the 1980s in order to minimize distribution of assembled vehicles to consumers throughout the
United
States (Rubenstein, 1992). Early in the twenty-first century, a second center of agglomeration for auto
production
emerged in central Mexico (Klier and Rubenstein, 2017).

The two key factors shaping the geography of ICE vehicle production have been agglomeration economics and
economies
of scale (for a more detailed explanation of these factors, see Klier and Rubenstein, 2015). Agglomeration
refers to
the association of productive activities in proximity to one another (Gregory et al., 2009, p. 14). According to
Ellison, Glaeser, and Kerr (2010, p. 1195), “the benefits of agglomeration ultimately reflect gains that occur
when
proximity reduces transport costs.” The motor vehicle industry is considered one of the most highly agglomerated
manufacturing industries (Ellison and Glaeser, 1997). Economies of scale refer to the fact that the volume of
output
can influence the per-unit cost of production (Cedillo-Campos, Sanchez-Garza, and Sanchez Ramirez, 2006; Klier
and
Rubenstein, 2015, p. 105; Lung, 2004; Truett and Truett, 1996, 2001, 2003; and Wynn-Williams, 2009). Note that
the
different steps in the production process of a light vehicle feature different economies of scale. In the case
of
ICE vehicle production, engine plants tend to operate at significantly higher volumes than final assembly
plants.
For example, in North America mean output in 2016 was 281,000 vehicles at ICE assembly plants and 421,000
engines at
engine plants (Klier and Rubenstein, 2022b, p. 9). As electrification of the auto industry expands, early
evidence
suggests that the economies of scale in battery production are comparable to those in engine production (see
ahead
to figures 22 and 23). We also know that engine plants and final assembly plants for ICE vehicle production tend
to
co-locate—that is, these two types of plants tend to be located close to each other (Klier and Rubenstein,
2021).
Early indications are that a similar pattern of co-location is emerging for battery plants and final assembly
plants
for EVs.

In 2023, more than one-half of all light vehicles produced in North America were assembled in auto alley, almost
equally divided between states north and south of the Ohio River.⁵
About one-fourth of these vehicles were
assembled
in Mexico (see figure 1). The remainder was divided between the Canadian portion of auto alley and locations in
the
United States outside of auto alley (see figures 1 and 2).

1. Light vehicle production in North America, by subregion, 2023

2. Locations of light vehicle assembly plants in North America, by production volume, 2023

Light vehicles with four types of propulsion systems are currently produced in North America (see Klier and
Rubenstein, 2022a, and the U.S. Department of Energy’s Alternative Fuels Data
Center for more details):

• Internal combustion engine vehicles, powered exclusively by a gasoline or diesel engine;
• Hybrid electric vehicles, powered
  primarily
  by a gasoline engine and by a small battery that may propel them in
  some circumstances, such as when starting up from a complete stop;
• Plug-in hybrid electric vehicles,
  powered by
  an electric motor connected to a battery, as well as an internal
  combustion engine; and
• Battery electric vehicles (also known as
  all-electric vehicles), powered exclusively by an electric battery.

In addition to these four types of vehicles, fuel cell
electric
vehicles (FCEVs) constitute a fifth type. FCEVs use a
propulsion system similar to that of electric vehicles. The fuel cell converts energy stored as hydrogen (stored
in
a tank on the vehicle) to electricity. No light FCEVs were built in North America in 2023. Around 14,400 were
expected to be built in North America in 2029, according to the S&P Global Mobility forecast data. We exclude
them
from our analysis because their share of total production in 2029 was expected to make up only 0.09%.

BEVs and PHEVs are commonly combined under the label “EV.” The terminology matters because these two types of
vehicles (along with FCEVs) are the ones classified as clean vehicles by the U.S. federal government, and
consequently, they may qualify for clean vehicle tax credits under the Inflation Reduction Act of 2022 (Pub. L. No.
117-169).⁶ PHEVs, HEVs, and ICE vehicles all contain
gasoline or diesel engines
and, therefore, are sometimes grouped together as gas-powered vehicles. Thus, within this article, depending on
the
information being conveyed, PHEVs are included along with BEVs as EVs or along with ICE vehicles and
HEVs
as
gas-powered vehicles.

In 2023, ICE vehicles made up 76.9% of the roughly 15.6 million light vehicles assembled in North America; HEVs
were
second in terms of the share of total volume assembled, at 13.5%—followed by BEVs, at 7.2%, and PHEVs, at 2.4%
(see
figure 3).

3. Share of total light vehicle production in North America for each vehicle type, 2023

What do we know about the production footprints for each of the four vehicle types across the subregions of
production within North America? It turns out that the footprints vary by vehicle type (as shown in figure 4):
In
2023, the United States had disproportionately large shares of BEV, HEV, and PHEV production in North America,
compared with its share of ICE vehicle production. In contrast, Canada and Mexico had larger shares of ICE
vehicle
production in North America, compared with their shares of production of the other vehicle types. So, for
instance,
assembly plants located in the United States were responsible for 89.9% of BEVs produced in North America,
compared
with only 61.5% of ICE vehicles. Conversely, Mexico’s assembly plants were responsible for 27.6% of ICE vehicles
produced in the region, but only 9.9% of BEVs.⁷

4. Volume and share of light vehicle production in North America, by subregion and vehicle type, 2023

Figures 5 and 6 show the distribution of assembly plants in North America by volume of production in 2023. Figure
5
displays the distribution of ICE vehicle and HEV production (the two types of vehicles powered exclusively or
primarily by a gas engine). In 2023, assembly plants in the U.S. portion of auto alley produced 51.4% of North
America’s ICE vehicles, as well as 72.8% of the entire region’s HEVs. Similarly, U.S. auto alley assembly plants
were responsible for 77.6% of North America’s PHEV production in 2023. However, U.S. auto alley assembly plants
produced only 28.8% of North America’s BEVs (all the percentages in this paragraph are from figure 4). Most BEVs
were made outside of auto alley, primarily in California (see figure 6). This pattern reflects Tesla’s large
share
of North America’s total BEV production in 2023.

5. Locations of ICE vehicle and HEV assembly plants in North America for domestic versus foreign automakers, by
production volume, 2023

6. Locations of BEV and PHEV assembly plants in North America for domestic versus foreign automakers, by
production volume, 2023

While studying auto alley within the United States, we elected to distinguish between assembly plants located
north
of the Ohio River and those located south of it. The three automakers with roots in the Detroit area dating back
more than a century—Ford, General Motors, and Stellantis⁸—have most
of their auto alley assembly plants located
in
the northern part of auto alley, whereas international automakers—rooted in Germany, Sweden, Japan, or South
Korea—have most of their assembly plants in the southern part of auto alley. Total production in 2023 amounted
to
4.2 million light vehicles in the northern portion of U.S. auto alley and 4.1 million vehicles in the southern
portion (see figure 1).

The northern and southern portions of U.S. auto alley (see note 5) both produced nearly identical shares of ICE
vehicles manufactured in North America in 2023, corresponding to about 3 million vehicles each (see figure 4 for
the
exact percentages). Within the United States, assembly plants in auto alley north accounted for a higher share
of
North America’s HEVs (39.2%) than the assembly plants in auto alley south did (33.6%) in 2023. That year the
northern portion of U.S. auto alley was responsible for slightly more production of two types of EVs (BEVs and
PHEVs) than the southern portion: At this early point in the industry’s shift toward electrification, assembly
plants in auto alley north produced 15.9% of North America’s total BEV production and 44.2% of its total PHEV
production, while plants in auto alley south produced 12.9% and 33.4%, respectively (see figure 4).

Future locations of ICE vehicle and BEV assembly plants

In this section, we describe the location of assembly plants for ICE vehicles and for BEVs (along with those for
HEVs
and for PHEVs) as forecasted through 2029. The data are based on forecasts developed by S&P Global Mobility for
the
assembly of light vehicles.⁹ The data are available to us at the
level of individual vehicle models produced at
each
assembly plant; we can distinguish propulsion systems by vehicle model. We recognize that specific numbers in
the
forecast can change over time. But here we are examining broad patterns concerning the auto industry’s
transition
toward electrification—as well as what they imply for its production footprint across North America.

As of December 2023, S&P Global Mobility forecasted an increase of 4.8% in overall light vehicle assembly across
North America over the next six years—from 15.6 million vehicles in 2023 to 16.4 million in 2029. The four
vehicle
types were expected by S&P Global Mobility to display different growth patterns between 2023 and 2029. Assembly
of
ICE vehicles was forecasted to decline by more than one-half, from 12 million to 5.2 million. In contrast, BEV
assembly during those six years was projected to increase more than sixfold, from 1.1 million to 7.3 million
(see
figure 7). As a result, the share of total light vehicle production in North America accounted for by ICE
vehicles
was anticipated to decline from around three-fourths in 2023 to less than one-third in 2029; in contrast, the
share
accounted for by BEVs was anticipated to increase from 7.2% in 2023 to 44.4% in 2029, surpassing ICE vehicle
output
by the end of the decade. The share of production of HEVs was also expected to grow substantially from 2023,
representing 20.7% of total light vehicle production by 2029 (again, see figure 7).

7. Change in the volume and share of light vehicle production in North America, by vehicle type, 2023–29

The mix of light vehicle types (by propulsion systems) in 2029 was forecasted by S&P Global Mobility to vary
across
subregions within North America (figure 8). As in 2023, the United States was expected to account for a
relatively
high share of North America’s total BEV assembly in 2029, although Canada and Mexico were expected to have
gained
shares of BEV production since 2023. In 2029, Mexico was expected to account for an even larger share of North
America’s ICE vehicle production than in 2023, and Canada for relatively larger shares of the region’s PHEV and
HEV
production (compare the subregions’ production share data in figures 4 and 8).

8. Volume and share of light vehicle production in North America, by subregion and vehicle type, 2029

Within North America, U.S. auto alley was expected to account for the same share of overall light vehicle
production
in 2029 as it did in 2023 (53.3%), according to our calculations using S&P Global Mobility data (again, compare
figures 4 and 8). However, both the northern and southern portions of auto alley were expected to increase their
respective shares of the region’s BEV production between 2023 and 2029; the share was forecasted to increase
slightly more rapidly in auto alley south (from 12.9% in 2023 to 25.3% in 2029) than in auto alley north (from
15.9%
in 2023 to 25.3% in 2029). Note that in 2029, North America’s BEV production was expected to be allocated almost
evenly among auto alley north, auto alley south, and the rest of the United States (each of these three
subregions
was forecasted to account for a little over one-quarter of BEV production in North America).

Figures 9 and 10 show the distribution of assembly plants across North America by the forecasted 2029 volumes of
ICE
vehicle and HEV production and of BEV and PHEV production, respectively. Comparing figures 5 and 9 shows how the
geography of ICE vehicle and HEV production—along with the output of these vehicle types at specific assembly
plants—is expected to change for domestic and international automakers between 2023 and 2029. Meanwhile,
comparing
figures 6 and 10 shows the same for BEV and PHEV production.

9. Locations of ICE vehicle and HEV assembly plants in North America for domestic versus foreign automakers, by
production volume, 2029

10. Locations of BEV and PHEV assembly plants in North America for domestic versus foreign automakers, by
production volume, 2029

The change in light vehicle production between 2023 and 2029 was not expected to be distributed evenly across the
subregions of North America (see figure 11). Over this span, light vehicle production was expected to increase
by
4.8% for the entire region. Only production in the United States was expected to grow more rapidly than that:
Light
vehicle output in the United States was anticipated to rise by 8.6% between 2023 and 2029. Production in Mexico
was
expected to grow modestly over this time, while production in Canada was expected to decline by more than 15%.
Within the United States, production in auto alley as a whole was forecasted to grow by 4.7%,¹⁰ with the northern
half
increasing more rapidly than the southern half (6.6% versus 2.8%). Light vehicle production within the United
States
outside of auto alley was forecasted to grow substantially between 2023 and 2029 (25.2%), almost all on account
of
the expected increase in production at Tesla’s facility in Austin, Texas.

11. Change in light vehicle production in North America, by subregion, 2023–29

Breaking down the data by vehicle type (that is, by propulsion technology), we note that the production of
BEVs,
PHEVs, and HEVs was forecasted to increase in all three North American countries in terms of vehicle units,
whereas
ICE vehicle assembly was forecasted to decline in all three countries (figure 12). In U.S. auto alley, the
expected
increase in BEV production was almost the same as the expected decrease in ICE vehicle production. The
increase
in
BEV assembly was forecasted to nearly make up for the decrease in ICE vehicle production in Mexico, but not
in
Canada. Overall, Canada was the only one of the three North American countries expected to show a decline in
light
vehicle production between 2023 and 2029.

12. Change in light vehicle production in North America, by subregion and vehicle type, 2023–29

Within the United States, 74% of the increase in BEV production between 2023 and 2029 was expected to be
clustered in
auto alley. Both halves of auto alley were expected to attract the same share of the growth in BEV assembly.
BEV
production outside of auto alley was forecasted to account for 26% of the increase.¹¹

Figure 12 suggests a massive change is expected in the mix of vehicle types produced in North America between
2023
and 2029. If final assembly plants were limited to producing vehicles with only one type of propulsion
technology,
the shift from ICE vehicle to BEV production indicated in figure 12 would require a substantial change in
the
industry: Automakers would have to add many new assembly plants that manufacture BEVs (around 25 plants) and
close
many assembly plants that manufacture ICE vehicles (around 26 plants).¹²

Yet, despite the expected substantial shift from ICE vehicle to BEV production, the footprint of assembly
plants
across North America was anticipated to remain virtually unchanged. Why? According to S&P Global Mobility,
the
production of BEVs was projected to take place—with a few exceptions—at existing light vehicle assembly
plants
that
were previously producing ICE vehicles. Hence, at the scale of North America as a whole, the transition from
ICE
vehicle to BEV production was expected to have a relatively minor impact on the industry’s vehicle assembly
footprint.

We explore this aspect of S&P Global Mobility’s auto outlook in more detail: Only seven new light vehicle
assembly
plants had been under construction or announced as of late 2023, and these were forecasted by S&P Global
Mobility to
open over the next six years from that point in time. The seven new plants will open in the following
locations:
two
in Georgia (by Hyundai and Rivian) and one each in North Carolina (by VinFast), Ohio (by Fisker), South
Carolina
(by
Volkswagen), Tennessee (by Ford), and Mexico (by Tesla) (see the green circles in figure 13). All seven were
forecasted to be utilized exclusively for BEV assembly. Only one assembly plant was forecasted to be
shuttered
by
2029—a joint venture plant by Mercedes-Benz and Nissan in Mexico (see the red x in figure 13).

13. Locations of light vehicle assembly plants in North America, by change in production volume, 2023–29

What about the fortunes of continuing assembly facilities across North America? According to our detailed
analysis,
of the 73 assembly plants open in 2023 and forecasted to still be open in 2029, 26 assembly plants
were
expected to
produce at least 10% more vehicles in 2029 than in 2023, 36 were expected to produce at least 10% fewer
vehicles,
and 11 were expected to experience little change, defined as producing less than 10% more or fewer vehicles
(figure
14).

14. Change in status of light vehicle assembly plants with an annual production volume of at least 25,000 units
in North America, by subregion, 2023–29

How can the rather stable number of assembly plants accommodate the dramatic change in the mix of vehicles
produced
over the medium term (as shown in figure 12)? S&P Global Mobility forecasted a greater diversity in the
product
mix
among existing vehicle assembly plants in 2029 relative to 2023. In 2023, 38 of the 74 light vehicle
assembly
plants
operating in North America produced vehicles with only one type of propulsion system, 24 produced vehicles
with
two
types, 11 produced vehicles with three types, and just one produced vehicles with four types. By 2029, the
number of
assembly plants producing vehicles with four types of propulsion systems was expected to have increased to
four
and
the number of plants producing vehicles with three types was expected to have increased to 21. Moreover, the
number
of assembly plants producing vehicles with only one type of propulsion system was forecasted to have
declined
from
38 in 2023 to 33 in 2029 (figure 15).

15. Light vehicle assembly plants in North America, by number of vehicle types produced, 2023 and 2029

Complementing figure 15, figure 16 reports the actual and projected numbers of light vehicle assembly plants
producing each vehicle type (ICE vehicle, HEV, PHEV, and BEV) in 2023 and 2029, respectively. Here each
vehicle
assembly plant is counted for each vehicle type it produces. We note that between 2023 and 2029, the number
of
plants assembling ICE vehicles was expected by S&P Global Mobility to drop from 65 to 45; yet the number of
plants
assembling BEVs was expected to jump from 18 in 2023 to 58 in 2029, more than offsetting that decline. The
number of
plants manufacturing HEVs was also projected to rise over this time, albeit more modestly, from 26 to 38. At
14,
the
number of plants assembling PHEVs was expected to be the same in 2029 as it had been in 2023.

16. Number of light vehicle assembly plants producing at least 1,000 units of each vehicle type in North America
and mean production at these plants, by vehicle type, 2023 and 2029

Let us return briefly to light vehicle assembly plants fully specializing in producing just one vehicle type
(see
again figure 15). Note that in 2023 there were 30 plants in North America that produced only ICE vehicles
and
eight
that produced only BEVs, according to the S&P Global Mobility data; the numbers were expected to be
virtually
the
opposite in 2029, with nine light vehicle assembly plants producing only ICE vehicles and 23 producing only
BEVs
(one plant was expected in 2029 to specialize in producing only HEVs). As the auto industry’s transition
toward
electrification progresses, it seems like the development and production of BEVs will be based on platforms
optimized for electric vehicles. This trend would support the growing number of vehicle assembly plants that
produce
BEVs only.¹³

Overall, the mean production per light vehicle assembly plant was forecasted to decrease from 211,000
vehicles in
2023 to 204,000 vehicles in 2029, according to our calculations using S&P Global Mobility data. In terms of
output
by vehicle type, the outlook varied: Consistent with the forecasted change in the product mix (figure 7),
mean
production was anticipated to decline substantially for ICE vehicles, but increase somewhat for HEVs and
PHEVs
and
substantially for BEVs (figure 16). Notably, the mean output of BEVs was expected to exceed that of ICE
vehicles
in
2029.

The distribution of engine and battery plants

When ICE vehicles dominated production, strong co-location between engine plants and final assembly plants
could
be
observed (see, for example, Klier and Rubenstein, 2021). Figure 17 shows that in 2023 both engine plants and
final
assembly plants were clustered in both auto alley and central Mexico. Will such a strong co-location between
final
assembly and propulsion system plants continue with the growth of electrification? We address this point in
two
parts: First, we examine whether strong co-location between light vehicle assembly plants and engine plants
will
continue; and second, we examine whether strong co-location between light vehicle assembly plants and
battery
plants
is likely to emerge over the next few years.¹⁴

17. Locations of assembly plants producing light vehicles with engines and locations of engine plants in
North
America, by production volume, 2023

Distribution of engine plants and assembly plants for light vehicles with gas engines

The shares of assembly operations for light vehicles with gas engines (ICE vehicles, HEVs, and PHEVs) among
Canada,
Mexico, and the United States (comprising the two halves of auto alley and the rest of the country) have
closely
matched the shares of engine production among these subregions (figure 18). For example, plants in Canada
assembled
10.6% of North America’s total output of light vehicles with engines and produced 9.5% of the engines
installed
in
light vehicles assembled within the region in 2023 (figure 18). That same year, plants in Mexico assembled
25.3%
of
the light vehicles with engines made in North America and produced 22.1% of the engines installed in light
vehicles
manufactured within the region. Similarly, the three areas within the United States—auto alley north, auto
alley
south, and the rest of the nation outside of auto alley—had roughly equivalent shares of North America’s
production
of light vehicles with engines and the region’s production of engines installed in such vehicles in 2023.
Notably,
in our analysis of the production of engines for light vehicles assembled in North America, we do not have
access to
data about engines produced within North America and exported outside of it (although we do have data on
engines
produced overseas and imported for installation in light vehicles assembled within the region).

18. Share of production of light vehicles with engines in North America and of engine production, by subregion,
2023 and 2029

Little change was forecasted for the distribution of production of light vehicles with engines across North
America
between 2023 and 2029. Similarly, the distribution of engine production across the region was anticipated to
not
change very much over this span. So, in 2029, the shares of assembly operations for light vehicles with
engines
among Canada, Mexico, and the United States (comprising auto alley north and south and the rest of the
nation)
fairly closely match the shares of engine production among these subregions—as they did in 2023. In 2029,
Canada
was
expected to have 10.3% of North America’s total assembly of light vehicles with engines and 4.9% of the
production
of engines installed in light vehicles manufactured in the region; Mexico was forecasted to have 27.6% of
North
America’s assembly of such vehicles and 28.4% of such engine production. U.S. auto alley’s share of the
assembly
of
light vehicles with engines in North America and its share of the production of engines installed in light
vehicles
made within the region were forecasted to be 55.5% and 54.5% in 2029, respectively. Most vehicles with
engines
and
most engines produced in North America were expected to continue to be made in U.S. auto alley—with the
shares
located in the northern half of auto alley running slightly ahead of the shares located in the southern
half.
For
U.S. auto alley as a whole, its share of the assembly of light vehicles with engines in North America and
its
share
of the engines produced for light vehicles made within the region were both expected to be slightly higher
in
2029
than in 2023 (see figure 18 and also compare figures 17 and 19).

19. Locations of assembly plants producing light vehicles with engines and locations of engine plants in
North
America, by production volume, 2029

How many engine plants are expected to be shut down as the production of ICE vehicles declines? At first
glance,
the
projected steep decrease in ICE vehicle production from 12 million vehicles in 2023 to 5.2 million in 2029
(figure
7) would suggest a drop in the need for engines equivalent to around ten fewer engine plants.¹⁵ However, S&P
Global
Mobility forecasted the closure of only five of the 34 full-sized engine plants in North America between
2023
and
2029—one each in Canada and Mexico and three in the northern portion of auto alley (see figure 20).

20. Number of engine plants in North America and mean production at these plants, by subregion, 2023 and 2029

At the same time, mean output at engine plants in North America was forecasted to decline by 24%—from about
384,000
in 2023 to about 291,000 units in 2029 (figure 20)—representing a much smaller drop (less than half) than
the
57%
decline anticipated in ICE vehicle production over this span (figure 7).¹⁶ The principal reason for this
relatively
modest decline in the volume of engine production is the projected 61% increase in the volume of HEV
production
between 2023 and 2029 (see figure 7).

Emerging distribution of battery plants

Lithium-ion battery production for EVs involves three main stages of manufacturing: cell, module, and pack.
Several
cells are put together into a module, with the number varying by individual manufacturer and automaker.
Similarly, a
pack contains a number of modules, which also vary by individual manufacturer and automaker (Klier and
Rubenstein,
2022a). Cells and packs may be manufactured in the same factory or in different factories, but typically,
according
to S&P Global Mobility, modules are produced in the same plants as cells and/or packs, rather than at a
different
set of plants. Therefore, we focus on cells and packs in this article.¹⁷

In 2023, nearly one-half of all battery cells installed in BEVs and PHEVs assembled in North America
were
imported
from other regions (that is, Europe and Asia), according to our analysis of the S&P Global Mobility
data.¹⁸ Only
10.3%
of cells installed in BEVs and PHEVs assembled in North America in 2029 were expected to be imported from
outside
the region (see figure 21).

21. Production share of battery cells installed in BEVs and PHEVs assembled in North America, by geography and
type of manufacturer, 2023 and 2029

Given the imported share of battery cells in 2023 just discussed, that means that 54.2% of cells installed in
BEVs
and PHEVs assembled in North America that year were made within the region. Of all the cells installed in
BEVs
and
PHEVs assembled in North America in 2023, 44.8% were manufactured at supplier-owned factories within the
region,
3.1% were manufactured at joint ventures in North America between suppliers and automakers, and 6.3% were
manufactured directly by an automaker from the region (Tesla). Three suppliers accounted for around
three-fourths of
all cells installed in BEVs and PHEVs manufactured in North America in 2023: 32.2% by Panasonic, 23.9% by
Samsung
SDI, and 17.5% by LG Energy Solution (Panasonic and LG Energy Solution produced some of their cells in North
America
and some of them outside the region) (figure 21).

With the size of individual cells being small, long-distance shipment of them might in principle be cost
effective,
as is the case with many engine parts. Most cell plants are owned either by independent suppliers—all of
which
are
based in East Asia—or by joint ventures between cell makers and automakers—in contrast with engine plants,
which
tend to be owned by automakers.¹⁹ These qualities of EV
battery
cell manufacturing raise the question of whether
the
emerging footprint of battery production will differ from the existing footprint of engine production, which
is
clustered in auto alley and central Mexico.

Of the eight plants in North America providing lithium-ion cells for BEVs and PHEVs assembled in the region
in
2023,
four were in the southern portion of U.S. auto alley, two in the northern portion of U.S. auto alley, two
elsewhere
in the United States, and none in Canada or Mexico. Whereas most automakers appear to be in the process of
converting existing assembly plants from ICE vehicle production to BEV production, we do not find examples
of
plants
currently producing gasoline engines being converted for EV battery cell production. Rather, cells are
produced
in
newly constructed factories.²⁰

S&P Global Mobility forecasted battery cell production from 28 factories in North America in 2029 (see
figures 22
and
23)—with some cells coming from overseas—for the 7.8 million BEVs and PHEVs to be assembled in the region
that
year
(figure 7). Three cell plants were expected to be operating in Canada in 2029 and 24 in the United States
(ten
in
auto alley south, eight in auto alley north, four elsewhere in the country, and two in locations yet to be
specified); Mexico was projected to have one cell plant in 2029 (see figure 23). Thus, cell plants were
forecasted
to reinforce the existing geography of light vehicle production clustered in auto alley. Note that auto
alley’s
shares in 2029 of cell production (55.7%) and of pack production (65.1%) were both expected to be above its
share of
overall light vehicle production (53.3%, as shown in figure 8). Furthermore, U.S. auto alley’s share of pack
production was anticipated to be noticeably higher than its share of engine production (54.5%, as shown in
figure
18). Plants in the United States outside of auto alley were expected to have greater shares of total cell
and
pack
production for BEVs and HEVs assembled in North America than plants in either Canada or Mexico in 2029 (see
figure
23); in addition, these plants in the United States were also projected to have a higher share of pack
production
(19.5%) than their share of engine production (4.7%, as shown in figure 18) that year.

22. Locations of battery cell and pack plants in North America, by production volume, 2029

23. Distribution of battery cell and pack plants in North America, by subregion, 2029

Also, note that when it comes to cell and pack production, the southern portion of U.S. auto alley was
expected
to
lead the northern portion in 2029 both in terms of plant count and production share (figure 23). Finally,
note
that
U.S. auto alley’s share of battery cell production (55.7%) was anticipated to be higher than its share of
BEV
production (50.6%, as shown in figure 8).

Meanwhile, the share of imported cells was forecasted to decline from 45.8% in 2023 to 10.3% in 2029 (figure
21).
Imports from both Europe and Asia were expected to decline considerably as a share of the total cell
production
volume used for BEVs and PHEVs assembled in North America between 2023 and 2029, according to our
calculations
using
S&P Global Mobility data.

LG Energy Solution was forecasted to become the largest supplier of EV battery cells in North America, with
30%
of
the market in 2029, primarily through a series of joint ventures with automakers. SK On was forecasted to
increase
its share of the EV battery cell market from 8% in 2023 to 15% in 2029, again primarily through joint
ventures
with
automakers. In contrast, Samsung SDI and Panasonic were forecasted to have reduced market shares in
2029—down to
9%
and 7%, respectively. Three automakers—Tesla, Toyota, and Volkswagen—were forecasted to produce 16% of cells
themselves rather than rely on joint ventures, as other automakers were expected to. Tesla has relied
primarily
on
Panasonic cells; the decrease in Panasonic’s market share reflects an expectation that Tesla will increase
production of its own cells in the same locations as its final assembly plants (see figure 21).

As S&P Global Mobility data show, 20 pack plants were already in operation in North America in 2023:
eight
in
the
northern portion of U.S. auto alley, seven in the southern portion of U.S. auto alley, four elsewhere in the
United
States, and one in Mexico. According to our calculations using these data, only 14% of packs were imported
into
North America in 2023, compared with 46% of cells. The earlier arrival of pack plants in North America
relative
to
cell plants is a reflection of how packs are manufactured. For the most part, packs are typically part of
the
chassis or undercarriage, now widely referred to as the vehicle’s platform. Consequently, packs are
manufactured
in
or near final assembly plants, as is the longstanding practice with ICE vehicle chassis.

Of the total number of packs forecasted to be installed in BEVs and PHEVs assembled in North America in 2029,
95%
of
those packs were expected to be put together by automakers alone, according to our calculations using S&P
Global
Mobility data. For the most part, these packs are manufactured in or near the automakers’ final assembly
plants.
Given that final assembly plants are clustered in auto alley, pack production will likely have a similar
geographical distribution in North America. S&P Global Mobility forecasted 28 plants to be producing battery
packs
in 2029: ten in the southern portion of U.S. auto alley, nine in the northern portion of U.S. auto alley,
five
elsewhere in the United States, and four in Mexico (refer to figure 23). Note that the concentration of
battery
pack
production in auto alley (65.1%) is significantly higher than that of BEV assembly in 2029 (50.6%, as shown
in
figure 8).

Co-location of propulsion system plants and final assembly plants

For each light vehicle assembly plant, S&P Global Mobility estimates the number of engines or batteries
(cells
and
packs) anticipated to be shipped from each propulsion system factory (that is, engine or battery plant). We
can
therefore calculate the median distance that each propulsion system travels between the location of its
production
and the final assembly plant where it is installed in a vehicle.

Many light vehicle assembly plants receive engines and/or batteries from more than one factory; in those
cases,
the
median distance is weighted by the share of engines and/or batteries an assembly plant is anticipated to
receive
from the various propulsion system plants. The median provides a more meaningful measure of distance than
the
mean
because a relatively small number of engines or batteries shipped from other regions of the world would
inflate
the
mean much more than the median and consequently result in a less accurate measure of the geographic
relationship
between propulsion system plants and final assembly plants.

In a previous study based on 2017 data (Klier and Rubenstein, 2021, p. 333), we calculated that in North
America
the
median distance between the places of production of engines and the final assembly plants where they are
installed
in vehicles was 376 miles (605 kilometers).²¹ We reported that
this distance, which is less than a one-day’s
drive,
represented a high degree of co-location. For this article, we have calculated that the median distance
between
engine production and final assembly was 272 miles (438 kilometers) in 2023 and is expected to be 222 miles
(357
kilometers) in 2029, using data from S&P Global Mobility. Thus, on average, the spatial relationship between
engine
plants and final vehicle assembly plants is even closer than what we found a few years ago—and this spatial
relationship is expected to be closer still in 2029.

For EVs (BEVs and PHEVs) assembled in North America, we have calculated two median distances for 2029: 1)
between
the
cell plants and the pack plants to which the cells are expected to be sent and 2) between the pack plants
and
the
final assembly plants to which the packs are expected to be sent.

We calculate that in 2029 the median distance between pack plants and final assembly plants in North America
is
expected to be 284 miles (457 kilometers). In other words, the median distance forecasted between battery
pack
plants and final assembly operations is rather similar to the median distance between engine plants and
final
assembly plants. Our analysis of S&P Global Mobility’s forecast data for 2029 indicates 56.0% of packs are
expected
to be made in final assembly plants and 36.1% in cell plants. We calculated the median distance between cell
plants
and pack plants to be only 40 miles (64 kilometers), dramatically lower than the median distance between
pack
plants
or engine plants and final assembly plants.

Thus, the footprints of cell and pack production are expected to align remarkably closely with the
distribution
of
final assembly plants forecasted to produce electrified vehicles (in particular, BEVs and PHEVs). U.S. auto
alley
was forecasted to have the bulk of the new EV production in 2029, whereas Canada and Mexico were expected to
have
relatively small shares.

Discussion

In this article, we compare the geographic footprint of the auto industry across North America in 2023 and
2029,
using actual and forecast data from S&P Global Mobility. At the scale of the integrated three-country
region, we
find surprisingly little change is expected in the geography of light vehicle production. Most BEVs are
forecasted
to be assembled in existing plants after they have been converted from ICE vehicle production. And most new
battery
(cell and pack) plants are being situated in auto alley, where engine plants have long been clustered.

This is partly explained by the fact that the underlying economic geography principles of motor vehicle
production—agglomeration economics and economies of scale—continue to influence location decisions for
production
even as the auto industry shifts toward electrification. The majority of plants assemble large volumes of
light
vehicles and are located within auto alley and central Mexico, resulting in a highly agglomerated industry
footprint. Battery cell plants, even though typically based in greenfield locations (that is, undeveloped
areas
in
and around cities), tend to be located reasonably close to battery pack plants, which are situated for the
most
part
close to final assembly plants. The most significant outlier from the clustering in auto alley—Tesla’s
choice of
California for its first BEV assembly plant—is still consistent with the broader auto industry’s goal of
minimizing
the shipping distances of assembled vehicles to consumers (and, therefore, the associated costs of doing so)
because
more than one-third of North America’s BEV sales through 2021 were in California (Klier and Rubenstein,
2015,
2022a).

Three factors are especially important in reaching the conclusion that only modest and selective changes are
likely
to occur in the distribution of light vehicle production within North America, at least through the end of
the
2020s. First, very few final assembly plants are forecasted to open or close in the years ahead, despite the
expectation that annual production in North America will increase by more than 6 million BEVs and decrease
by
more
than 6 million ICE vehicles (see figure 7). The reason is that with only a few exceptions, automakers have
begun
converting existing final assembly plants to accommodate BEV production. With very few plant openings and
closures,
the overall footprint of assembly operations in North America is therefore unlikely to change.

Second, although propulsion system plants are typically not being converted from engine to battery
production,
new
battery plants are being constructed in similar locations as those of existing engine plants. The location
of
battery plants—like that of engine plants—is strongly influenced by the desire for co-location with light
vehicle
assembly plants. Thus, if the distribution of assembly plants is not changing, new battery plants are likely
to
exhibit a similar spatial pattern as that of the existing engine plants.

Third, the growth of hybrid vehicles also suggests that there will be rather minor differences between the
footprints
of BEV and ICE vehicle assembly and between the distribution of battery and internal combustion engine
production.
S&P Global Mobility forecasted HEVs and PHEVs jointly to account for 24.0% of North America’s total vehicle
output
in 2029 (see figure 7). Furthermore, 47 of the 80 assembly plants in operation in 2029 were expected to be
producing
vehicles with more than one of the four main propulsion systems and the number of assembly plants producing
vehicles
with at least three different types of propulsion systems was forecasted to increase from 12 in 2023 to 25
in
2029
(see figure 15). The ability to produce vehicles with a variety of propulsion systems at a single assembly
plant
reinforces the forecast that few assembly plants will be opened or closed in the years ahead. At the same
time,
it
is expected that by 2029 a much larger number of assembly plants (nearly a third) will be specializing in
the
production of electric vehicles (as we mentioned earlier, 23 of the 80 final assembly plants in 2029 were
projected
to produce only BEVs).

The growth of hybrid production is even more significant for reducing the disruption to propulsion system
plants
over
the medium term. The dramatic decrease in the production of ICE vehicles (over 6 million units)—along with
the
commensurate decline in demand for engines—between 2023 and 2029 is not expected to result in a massive
number
of
engine plant closures. One of the main reasons for this is that HEVs also require engines. Hence, the
forecasted
decline in demand for engines for ICE vehicles would be offset to a considerable extent by the growth in
demand
for
engines for HEVs, as production of that type of vehicle was also expected to grow (see figure 12).²²

We started this article with the question of whether the transition to electrified vehicles will result in
material
changes to the automotive industry’s production footprint over the medium term. We had expected to focus on
similarities and differences between the longstanding distribution of ICE vehicle production on the one hand
and
the
emerging distribution of BEV production on the other. What we learned suggests that there will be rather
minimal
changes to the industry’s footprint over the medium term. Existing vehicle assembly plants will be able to
accommodate the production of vehicles featuring propulsion systems other than the internal combustion
engine.
Furthermore, increased production of hybrid vehicles containing both gas engines and batteries will likely
further
limit changes in the geography of motor vehicle production across North America.

It is important to mention, however, that the degree of change in the industry footprint depends on the scale
of
analysis. Auto alley is expected to see few challenges to its dominant role in light vehicle production, yet
at
the
local scale of specific cities and towns, the opening of battery plants—plus the possible closure of engine
plants
or reduced production activity at them—may well lead to changes in local economic fortunes.

Notes

¹ The U.S. Department of Energy’s Alternative Fuels Data Center
states the following: “An EV is defined as a vehicle
that can be powered by an electric motor that draws electricity from a battery and is capable of being
charged
from
an external source. An EV includes both a vehicle that can only be powered by an electric motor that draws
electricity from a battery (all-electric vehicle) and a vehicle that can be powered by an electric motor
that
draws
electricity from a battery and by an internal combustion engine (plug-in hybrid electric vehicle).”

² See also Klier and Rubenstein (2024).

³ Note that we have used data from the same source in previous
work;
see, for example, Klier and Rubenstein (2017) and
Brincks, Klier, and Rubenstein (2016). In those earlier articles, the data source was listed as IHS Markit
Automotive Solutions, but it is now called S&P Global Mobility as a result of the merger of S&P Global and IHS
Markit in February 2022.

⁴ In this article, we account for only EV batteries and gas
engines
in our analysis of the anticipated shifts in light
vehicle propulsion systems and their impacts on the auto industry footprint across North America. We do not
consider
transmissions or components of the electrified powertrain in this analysis.

⁵ U.S. auto alley north consists of Illinois, Indiana, Michigan,
and
Ohio. U.S. auto alley south consists of Alabama,
Georgia, Kentucky, Mississippi, North Carolina, South Carolina, and Tennessee. Southwestern Ontario, Canada,
is
sometimes considered as an extension of auto alley. However, in our analysis (as reflected throughout our
figures),
southwestern Ontario is treated separately in order to more clearly identify expected changes in production
in
each
of the three light-vehicle-producing countries of North America.

⁶ The U.S. Department of Energy provides further details about
the
clean vehicle tax credits on this Alternative Fuels
Data Center webpage and this
Office
of Energy Saver webpage.

⁷ Figure 4—as well as other figures within this article organized
by
light vehicle type (that is, by the propulsion
system for the vehicle)—order the four types of vehicles (noted in the column headings) from electric-only
to
hybrid
to gas-only. That way, readers can group neighboring columns when considering plug-in vehicles as EVs (BEVs
and
PHEVs) or as vehicles with gas or diesel engines (PHEVs, HEVs, and ICE vehicles).

⁸ Stellantis N.V. was formed in 2021 through the merger of Fiat
Chrysler Automobiles N.V. (FCA) and Peugeot S.A. FCA
operated with two main subsidiaries—FCA Italy (previously Fiat) with headquarters in Italy and FCA US
(previously
Chrysler, one of the original Big Three automakers) with headquarters in Michigan (Piovaccari, 2021; Bond,
2020;
and
Vellequette, 2014). We classify Stellantis as a Detroit-based company (that is, a domestic automaker) in
this
article because all the brands of vehicles it assembles in North America stem from Chrysler.

⁹ We base our analysis on a production forecast issued by S&P
Global
Mobility. This outlook extended through 2029 at
the time of our research; it is updated at a monthly frequency. Specific quantities and product allocations
can
vary
from month to month. After comparing several monthly forecasts from 2023, we found our conclusions to be
robust.
The
forecast data for 2029 represent information issued by S&P Global Mobility in December 2023. (The actual
data
for
2023 represent information issued by S&P Global Mobility in January 2024.)

¹⁰ This production growth value for auto alley as a whole is
not
shown in figure 11; it is based on our calculations
using the same data source listed in figure 11.

¹¹ We derived the share of 2023–29 growth in BEV output within
the
United States for the entirety of auto alley
mentioned in this paragraph by using the data for its two halves (north and south), as well as the whole
country,
reported in figure 12. The share of this growth outside of auto alley was also derived from data reported in
figure
12.

¹² The rule of thumb for annual production on a final assembly
line
is around 250,000 units (which is based on a rate of
one vehicle per minute during two eight-hour shifts per day, for five days a week, over 50 weeks per year).

¹³ According to an Automotive News Europe article
summarizing an interview with Pedro Pacheco, vice president of
research at Gartner, “dedicated EV platforms have given automakers the freedom to design assembly lines that
are
suited to their characteristics, including a smaller powertrain and flat battery floor” (Sigal, 2024).

¹⁴ Note that we discuss the location of battery production only
for
BEVs and PHEVs, that is, vehicle types with a
battery that can be charged directly (with a plug). We do not cover the geography of HEV battery production
in
our
analysis. In terms of driving on electricity alone, HEV batteries are designed to provide a shorter range
than
either BEV or PHEV batteries. HEV batteries are typically much smaller than BEV or PHEV batteries and differ
from
them in other ways (for details, see Csere, 2021).

¹⁵ This estimate of ten fewer engine plants is based on our
calculation for the mean production at engine plants in
North America (Klier and Rubenstein, 2022b).

¹⁶ Based on our calculations using data reported in figure 7,
the
production of vehicles with engines (ICE vehicles,
HEVs, and PHEVs) was expected to decline by 37% between 2023 and 2029. Generally speaking, in this article,
we
discuss engine production during the unfolding transition toward electrification. The first-order
relationship
we
focus on is that for every fully electric vehicle, one fewer internal combustion engine will be needed. Note
that
there are possible second-order adjustments at play (which are beyond the scope of this article). For
example,
there
could be changes to the engine mix produced at engine plants (as the range delivered by batteries for
hybrids
increases, the size of the engine might shrink), and these plants might start to produce components of the
electrified powertrain.

¹⁷ Note that the majority of electric batteries for vehicles
currently follow the cell-module-pack production sequence
(and are forecasted to do so in 2029); however, there are alternative approaches, such as skipping the
module as
a
separate production stage or integrating the modules (or cells) into the vehicle body.

¹⁸ Note that the units of cells we report have been normalized
to
the number of BEVs and PHEVs assembled in North
America; that is, we count the multitude of cells installed in a single BEV or PHEV (averaging over 700
cells
per
light vehicle, according to our calculation based on S&P Global Mobility data) as one unit of cell
production
and
report the total cell production volumes accordingly.

¹⁹ S&P Global Mobility forecasted Toyota and Volkswagen would
be
producing some of their own cells for EV batteries by
2029, as Tesla had already been doing in 2023, as shown in figure 21.

²⁰ Plants producing EV batteries, including their cells, have
unique requirements regarding “production flow, material
handling, environmental control and fire safety”; such factories also need to be able to accommodate very
large
electrical loads (two to three times the loads of a typical light vehicle assembly plant) (Verner, 2023). On
balance, these and other factory design factors suggest that EV battery manufacturing plants are likely to
be
built
on greenfield sites (that is, undeveloped areas in or around cities), in lieu of transforming existing
engine
plants
for EV battery production.

²¹ We calculated the distances as straight-line distances.

²² While the production of PHEVs (which also require engines)
was
expected to grow between 2023 and 2029 (figure 12),
the PHEV share of total light vehicle production in North America was projected to decrease (see figure 4
versus
figure 8); thus, PHEVs likely won’t contribute much to helping offset the decrease in demand for engines due
to
ICE
vehicle production reduction. Note also that we did not explore possible changes to engine characteristics,
along
with related supply chain impacts, during the auto industry’s transition to electrified vehicles—to the
extent
that
hybrid vehicles (HEVs and PHEVs) replace ICE vehicles.

References

Bond, Vince, Jr., 2020, “Newly merged FCA-PSA entity will be called Stellantis,” Automotive
News,
July 15,
available
online by
subscription.

Brincks, Corey, Thomas Klier, and James M. Rubenstein, 2016, “The role of national champions in the evolving
footprint of vehicle production in Europe—1990–2013,” International Journal of Automotive Technology
and
Management,
Vol. 16, No. 2, pp. 130–146. Crossref

Cedillo-Campos, Miguel Gaston, Juan Sanchez-Garza, and Cuauhtemoc Sanchez Ramirez, 2006, “The new relational
schemas
of inter-firms cooperation: The case of the Coahuila automobile cluster in Mexico,” International
Journal
of
Automotive Technology and Management, Vol. 6, No. 4, pp. 406–418. Crossref

Csere, Csaba, 2021, “Battery taxonomy: The differences between hybrid and EV batteries,” Car and
Driver, August 10,
available
online.

Ellison, Glenn, and Edward L. Glaeser, 1997, “Geographic concentration in U.S. manufacturing industries: A
dartboard
approach,” Journal of Political Economy, Vol. 105, No. 5, October, pp. 889–927. Crossref

Ellison, Glenn, Edward L. Glaeser, and William R. Kerr, 2010, “What causes industry agglomeration? Evidence
from
coagglomeration patterns,” American Economic Review, Vol. 100, No. 3, June, pp. 1195–1213. Crossref

Gregory, Derek, Ron Johnston, Geraldine Pratt, Michael J. Watts, and Sarah Whatmore (eds.), 2009, The
Dictionary of
Human Geography, 5th ed., Chichester, UK: Wiley-Blackwell.

Klier, Thomas H., and James M. Rubenstein, 2024, “From ICE to BEV production in Europe: Industry geography in
transition,” International Journal of Automotive Technology and Management, Vol. 24, No. 1, pp.
95–117. Crossref

Klier, Thomas H., and James M. Rubenstein, 2022a, “North America’s rapidly growing electric vehicle market:
Implications for the geography of automotive production,” Economic Perspectives, Federal
Reserve
Bank of Chicago,
No. 5. Crossref

Klier, Thomas, and James M. Rubenstein, 2022b, “Trade agreements and the geography of motor vehicle
production in
North America and Europe,” International Journal of Automotive Technology and Management, Vol.
22,
No. 1, pp. 4–28.
Crossref

Klier, Thomas, and James M. Rubenstein, 2021, “ICE age geography: Powertrain sourcing in Europe and North
America,”
International Journal of Automotive Technology and Management, Vol. 21, No. 4, pp. 322–342. Crossref

Klier, Thomas H., and James M. Rubenstein, 2017, “Mexico’s growing role in the auto industry under NAFTA: Who
makes
what and what goes where,” Economic Perspectives, Federal Reserve Bank of Chicago, Vol. 41, No.
6,
available online.

Klier, Thomas H., and James M. Rubenstein, 2015, “Auto production footprints: Comparing Europe and North
America,”
Economic Perspectives, Federal Reserve Bank of Chicago, Vol. 39, Fourth Quarter, pp. 101–119,
available
online.

Lung, Yannick, 2004, “The changing geography of the European automobile system,” International Journal
of
Automotive
Technology and Management, Vol. 4, Nos. 2–3, pp. 137–165. Crossref

Piovaccari, Giulio, 2021, “After long journey, Fiat Chrysler and PSA seal merger to become Stellantis,”
Reuters,
January 16, available online.

Rubenstein, James M., 1992, The Changing US Auto Industry: A Geographical Analysis, London and
New
York: Routledge.

Sigal, Peter, 2024, “EVs will be cheaper to build than ICE cars by 2027, analyst says,” Automotive News
Europe, March
8, available
online by subscription.

Truett, Lila J., and Dale B. Truett, 2003, “The Italian automotive industry and economies of scale,”
Contemporary
Economic Policy, Vol. 21, No. 3, July, pp. 329–337. Crossref

Truett, Lila J., and Dale B. Truett, 2001, “The Spanish automotive industry: Scale economies and input
relationships,” Applied Economics, Vol. 33, No. 12, pp. 1503–1513. Crossref

Truett, Lila J., and Dale B. Truett, 1996, “Economies of scale in the Mexican automotive sector,”
Journal
of
Productivity Analysis, Vol. 7, No. 4, October, pp. 429–446. Crossref

Vellequette, Larry P., 2014, “Fiat’s European holding company now called FCA Italy,” Automotive News
Europe, December
16, available
online by subscription.

Verner, David, 2023, “Battery plant design do’s and don’ts,” Assembly Magazine, September 15, available online by
subscription.

Wynn-Williams, Michael S., 2009, “Economies of scale in the automobile industry,” in Surfing the Global
Tide:
Automotive Giants and How to Survive Them, Basingstoke, UK: Palgrave Macmillan, pp. 41–71.

Opinions expressed in this article are those of the author(s) and do not necessarily reflect
the views of the Federal Reserve Bank of Chicago or the Federal Reserve System.