Evaluating the Macroeconomic Impacts of Broadband Funding from the Bipartisan Infrastructure Law

Table of Links

Abstract and Introduction

2. Literature Review

2.1 Reviewing Broadband Infrastructure’s Impact on the Economy

2.2 Previous Research into IO Modeling of Broadband Investment

2.3 Context of the Bipartisan Infrastructure Act through Previous Research

3. Methods and 3.1 Leontief Input-Output (IO) Modeling

3.2 Ghosh Supply-Side Assessment Methods for Infrastructure

3.3 Data and Application

4. Results and 4.1 To what extent does the Bipartisan Infrastructure Law allocate funding to unconnected communities in need?

4.2 What are the GDP impacts of the three funding programs within the Bipartisan Infrastructure Law?

4.3 How are the supply chain linkages affected by allocations from the Bipartisan Infrastructure Law?

5. Discussion

5.1 To what extent does the Bipartisan Infrastructure Law allocate funding to unconnected communities in need?

5.2 What are the GDP impacts of the three funding programs within the Bipartisan Infrastructure Law?

5.3 How are supply chain linkages affected by allocations from the Bipartisan Infrastructure Law?"

Conclusion

Acknowledgements and References

6 Conclusion

In this paper, we assessed the macroeconomic impacts of the Broadband, Equity, Access and Deployment program, American Connectivity Program, and Tribal Broadband Connectivity Program. We firstly evaluated the program allocation and enrollment on a state-by-state basis, Secondly, we estimated the national GDP impact by using national accounting data to report the downstream impacts via the Ghosh Supply-side model and upstream impacts via the Leontief Demand-side model. To our knowledge, this is the first macroeconomic assessment of the broadband infrastructure programs within the Bipartisan Infrastructure Law.

In general, we found that higher BEAD allocations target areas with poor current broadband supply. While this may be logical, as it illustrates funding going to areas most in need, this could not have been assumed a priori given politically-motivated funding is not always rationally allocated. We also found some outliers with BEAD allocations, for example, with Alaska receiving roughly US $39,000 per unconnected household. Moreover, Louisiana ACP enrollment is twice as high as the mean state enrollment, with around 26.6% of homes enrolling (versus a mean of 13.3%).

In terms of macroeconomic impact, the total direct contribution to US GDP by the program could be as high as US $84.8 billion, $55.2 billion, and $5.99 billion for the BEAD program, ACP, and TBCP, respectively. Overall, the broadband allocations could expand US GDP by $146 billion, with a multiplier of 2.45. The NAICS industries that may benefit the most from these programs include the Information sector (NAICS 51, $16.9 billion), the Professional, Scientific, and Technical Services Sector (NAICS 54, $11.9 billion), and the Manufacturing Sector (NAICS 31-33, $6.85 billion). Insights about the supply-chain industry were also evaluated, including the relative interdependencies of sectors on broadband telecommunications and how certain sectors were more reliant on high-speed, reliable Internet as a value-added input good than other sectors (Agriculture, forestry, fishing, and farming; NAICS 11; 84.6% downstream and Retail Trade; NAICS 44-45; 99.8% downstream).

As with any method, there are relevant limitations which are worth highlighting. Firstly, the IO approach in this paper is static and does not account for potential dynamic impacts in the macroeconomy as supply and demand forces play out in a market environment. Thus, we do not account for structural changes or how measured stimulus would alter sectoral interdependencies. Like other IO approaches, we assume constant proportionality of industry inputs, so there are no changes to the original technical coefficients matrix regardless of sectoral augmentation from additional output, increased productivity, or novel technologies (Galbusera & Giannopoulos, 2018; Miernyk, 1966). Moreover, this approach depends on constant returns to scale, with no decreasing marginal cost of production as output increases. Importantly, this IO model approach produces the upper bound on GDP impacts, so growth could be lower than projected (Jones, 1968; Oosterhaven, 1988; Walheer, 2020). One real benefit of an IO approach is understanding potential inter-sectoral interactions, which has driven the focus of this paper to supply linkage analysis.

Future research could expand the methodology in this paper to focus on dynamic macroeconomic modeling strategies to study how spending increases from building new broadband infrastructure compares to the estimates produced here. Additionally, future studies could examine how to better allocate funding towards future infrastructure investments, ideally maximizing the effectiveness of allocations through the BEAD and ACP enrollment programs.