Logo mobile

Purdue Center for
Regional Development

An X-Ray of Broadband Access in the Upper Midwest

Posted by Roberto Gallardo on June 24, 2019, in Broadband.

Written by Roberto Gallardo and Bo Beaulieu.

In a recent broadband-related post, we mapped the 25 Megabits per second (Mbps) download and 3 Mbps upload (25/3, for short) fixed broadband footprint in Indiana. The Federal Communications Commission (FCC) employs this speed threshold to define broadband. Because of this analysis, we found that as of 2017, around 10 percent of housing units in Indiana did not have access to 25/3, which equates to roughly 673,000 Hoosiers. Furthermore, we noted that 62 percent of housing units in completely rural areas had access to 25/3 versus 99 percent in completely urban areas.

For this post, we undertook a broader geographic (included Illinois, Indiana, Michigan, Minnesota, Ohio, and Wisconsin) and temporal (2014 and 2017) look at the 25/3 broadband footprint and explored the broadband footprint of Census block groups located in completely urban, mostly urban, mostly rural, and completely rural areas. Completely urban areas had more than 99 percent of their population living in urbanized areas; mostly urban had between 50 and 99 percent living in urbanized areas; mostly rural between 1 and 49 percent living in urbanized areas; and completely rural had less than 1 percent living in urbanized areas.

We have included, in addition to the 25/3 metric, a symmetrical 25/25 Mbps and the ultra-fast 1/1 Gigabit per second (Gbps) speed thresholds as of 2017 given that an increasing number of homes and businesses are involved in producing or uploading, not only consuming or downloading information and applications. This means that upload speeds, aside from download speeds, are becoming more critical for community economic development.

Data for this analysis were secured from the FCC Form 477. For purposes of full disclosure, this dataset is known to be somewhat problematic given that it is likely to overestimate broadband access. Some of the issues associated with this dataset include: 1) it reports advertised—not actual—speeds; 2) data are self-reported by carriers and have not been validated; and 3) the unit of analysis is Census blocks and as such, if any structure in that block has access, the entire block is identified as served. Despite its limitation, it is the only nationwide dataset available.

Figure 1 shows the percent of housing units by state inside the 25/3 footprint as of 2014 and 2017. It is clear that the footprint increased in all states between 2014 and 2017, with Ohio having the highest share inside the 25/3 footprint in 2017 with 95 percent coverage.

Figure 1

Figures 2 & 3 highlight 2014 and 2017 25/3 broadband access by state and neighborhood type. Again, the 25/3 broadband footprint improved between 2014 and 2017 across all states and neighborhood types. For example, slightly more than one-third of housing units (35.2 percent) in completely rural neighborhoods in Illinois had access to 25/3 in 2014 increasing to more than half (55.4 percent) by 2017.

Figure 2
Figure 3

However, a gap remains between completely urban and completely rural areas, despite the significant improvements witnessed in rural areas. For example, as of 2017, virtually all housing units in Wisconsin’s completely urban neighborhoods (99.9 percent) had access to 25/3 compared while less than two-thirds of housing units in completely rural areas (63.3 percent) had access to this level of service.

Among the six states, Illinois had the lowest share of housing units in completely rural areas (55.4 percent) with access to 25/3 in 2017, followed by Indiana (62 percent). Minnesota, on the other hand, had the largest share of completely rural areas with access to 25/3 with nearly 80 percent, followed by Ohio with 72.5 percent. In fact, Minnesota had the largest percentage point increase (27.1) in the share of completely rural areas with access to 25/3 between 2014 and 2017.

Figure 4 shows the 2017 25/3 footprint of the upper Midwest region we analyzed. Gray areas specify where 25/3 is available while orange areas represent where 25/3 is not available. A darker orange indicates a higher number of housing units in that particular block. Crosshatched areas denote urban blocks.

Figure 4. 25/3 Broadband Footprint in the upper Midwest, 2017

Figure 4 Source: FCC Form 477 December 2017-v2

When all is said and done, as of 2017, about 1.7 million housing units in these six upper Midwest states did not have access to 25/3 affecting 3.5 million residents. Michigan had the highest share (25.5 percent) of housing units in the region with no 25/3 access, followed by Illinois (17.6 percent) and Wisconsin (17.2 percent). Minnesota had the lowest share, with less than 10 percent of the total.

However, broadband access changes significantly when analyzing two additional speed thresholds: 25/25 Mbps and 1/1 Gbps. Figures 5 & 6 show the broadband footprint in the six-state region using a symmetrical 25/25 Mbps speed. Notice how orange areas (with no access to this speed) across the six-state regions increased when using this speed threshold.

Figure 5. 25/25 Broadband Footprint in the Upper Midwest, 2017

Figure 5 Source: FCC Form 477 December 2017-v2

Regarding neighborhood types, Figure 6 showcases the percent of housing units with access to 25/25 Mbps in 2017. Again, a higher share of completely urban neighborhoods had access to symmetrical 25/25 Mbps across the region compared to completely rural. In Illinois, for example, 92.9 percent of housing units in completely urban neighborhoods had access to 25/25 Mbps compared to only 39.9 percent in completely rural neighborhoods, a 53 percentage points difference. Notice that Minnesota had the highest share of completely rural areas with access to 25/25 at 53.4 percent, followed by Illinois (39.9 percent) and Indiana (38.7 percent).

Figure 6

Lastly, the ultra-fast speed threshold of 1/1 Gbps footprint is shown in Figure 7. Again, orange areas (with no access to this speed) increased significantly across the region with virtually all of Illinois, Ohio, and Wisconsin without access to this speed.

Figure 7. 1/1 Gbps Broadband Footprint in the Upper Midwest, 2017

Figure 7 Source: FCC Form 477 December 2017-v2

When examining data by neighborhood types, it is clear that the percent of housing units with access to 1/1 Gbps decreased significantly (see Figure 8). While the gap between completely urban and completely rural exists, it is not as wide as with the other much slower speed thresholds. Minnesota, again, outperforms all other states in the region with one-fifth of housing units in completely urban and also completely rural with access to this speed threshold.

Figure 8

As of 2017, the six-state region had 23.1 million housing units, or about 52.4 million residents. Of these, approximately 10.8 million housing units (46.9 percent), or about 23.7 million residents, lacked access to symmetrical 25/25 Mbps broadband. Less than one-third (30 percent) were located in Ohio followed by Michigan with 28.3 percent. On the other hand, Minnesota had the lowest share of the region’s total with 5.3 percent without access to 25/25 Mbps.

Regarding the 1/1 Gbps speed, the percent of housing units without access to this speed threshold skyrocketed. About 22.4 million housing units (96.9 percent) in the region did not have access to this speed threshold, 10 percentage points higher than the US share of 86.4 percent. While 1/1 Gbps represent ultra-fast connectivity, it provides ample room to accommodate the increasing number of devices per housing unit, businesses, and farms connecting to the internet while comfortably handling future internet applications that require faster and faster connections.

While access to 25/3 has improved between 2014 and 2017 in the region, there still is a serious gap between urban and rural areas. As faster, symmetrical speed thresholds are analyzed, the share of housing units with access to this type of coverage decreases. Efforts are needed to ensure symmetrical speeds are improved so that the region can remain competitive. Without question, digital capacity will be an essential requirement for any state or community that wishes to be competitive in a global marketplace.

In closing, it is important to remember that our dataset is likely to overestimate broadband access. As such, it is important for local leaders, residents and providers to validate the information, whenever possible. Moreover, this analysis focused on broadband access only, which does not allow us to examine the broader digital inclusion picture. Digital inclusion is the meaningful use of broadband applications that improve the quality of life and is affected strongly not only by availability and access, but also by devices, internet subscription types, and socioeconomic factors.

Our next posts will look at more robust digital inclusion metrics—digital distress and the digital divide index—in the region, providing a broader understanding of digital inclusion and its community economic and workforce development implications.