TPRC 2025: The 53nd Research Conference on Communications, Information and Inter

Link to paper

Abstract:
The FCC Broadband Data Collection (BDC) program has had—and will continue to have—tremendous impact on directing policy interventions and funding towards the goal of achieving broadband equity, access, and deployment across the United States.
In this paper, we share our experience analyzing the data disseminated by the FCC as part of this program.
We focus on discussing the challenges and limitations that one may encounter when exploring the datasets made publicly available as part of this program.
Examples are the lack of direct, public data on the fabric layer; the retroactive removal of availability records from past data releases; and the purely file-based data serving model.
We provide recommendations to stakeholders on ways to overcome these challenges and cope with limitations.
These recommendations seek to introduce best practices for processing and analyzing the BDC data.
Where appropriate, we also bring suggestions to the FCC on approaches to eliminate data limitations and lower barriers to analysis.
These suggestions involve changes to how BDC data is published, served, updated, and summarized by the FCC.

Link to Paper

Abstract:
This paper presents initial research to contribute to the discussion surrounding broadband access data. In this paper, an approach to quantifying community behaviors across critical demographics for broadband mapping is explored and a research roadmap presented. Ultimately, the goal of this research is to offer a recommendation for improvement of the U.S. federal broadband map data process to establish the best holistic representation of broadband access for public policy research and decision making. The paper explores two initial research questions. First, how have community behavior and requirements data been collected and considered in U.S. broadband development initiatives, subsidies, and policies? Second, how have community behaviors been quantified in previous policy research? Finally, this paper presents a research roadmap on how to identify and apply community behavior data to the National Broadband Map for effective policy decisions.

Link to paper

Abstract:
The South Korea telecom services market reports annual revenues of $32 billion. The country is noted for market-based network usage fees in which content providers (both domestic and foreign) negotiate access to broadband networks. The magnitude and volume of such fees are not public, but news reports suggest they amount to less than 1 percent of the total market revenue. This paper investigates the size and volume of the network usage fee regime and impacts on related broadband and content markets. 

As usage fees have been place for some years in Korea, this paper reviews the development of the Korean broadband and content markets and whether changes can be observed after the introduction of usage fees to the industry structure, revenue, market size, concentration, penetration, technological development, and other variables. International comparisons are offered to see whether there are statistically significant differences in countries without formal usage fee regimes. 

Results are expected to be correlative, not causative. Theoretical modeling is explored in parallel to explore whether and how results could change under different circumstances.

A conceptual framework suggests expected relationship between variables and defines objectives for the research process. In this respect, the authors hypothesize that network usage fees have a neutral effect on the market, or at best positive. Alternative scenarios are explored, for example the refusal to supply by content providers and the lack of reinvestment of fees into the network. 

The paper explores other reasons why the Korean regime may be an appropriate model or not, for example, the discrete number of broadband providers, the presence of an authority to meter and publicize traffic levels, the ease of measurement of international traffic because of limited entry points to the country, policy orientation towards technological innovation, and so on. 

The paper is relevant for at least 4 themes promoted at TPRC52 including
• Broadband Deployment, Access and the Digital Divide, Criticality of Infrastructure
• Social Dimensions to Communications and Internet Policy
• Policies Governing the Digital Economy
• Geopolitics, Digital Sovereignty, and Tech Regulation.
Moreover, it offers new, original, and valuable data and analysis for policy scholars including a bibliography of Korean sources of information on usage fees translated and summarized for an English-speaking audience, a formal presentation of Korean content market development over the last decade and reference to international proceedings on the similar topic in other countries as appropriate. 

The paper offers a timely review of the small but emerging literature on the topic of network usage fees and interconnection and the significant literature in Korean on the nation’s broadband policy, which is not necessarily accessed by English language scholars. 
The work is significant because many nations consider whether and how to engage in such regimes and whether market-based or regulated approaches are superior. Proceedings have been undertaken or underway in USA, European Union, Brazil, among other regions.

Link to paper
Abstract:
Device affordability is one of the key barriers to Internet access and use in most low-and-middle income countries. Reducing and/or eliminating taxes on devices has emerged as a major way to make devices more affordable. However, the empirical evidence on the impact of reducing and/or eliminating tax on device access remains limited. To help fill this gap, this study estimates the impact of Colombia’s 2017 VAT exemption on smartphone penetration rates. Using the Synthetic Control Method (SCM) applied to country-level panel data from 2009 to 2021, our study finds that smartphone penetration in Colombia reached 66.8 percent in 2021. This represents an increase of 7.6 percentage points over the 59.2 percent level of its synthetic counterpart, equivalent to a 12.8 percent relative increase. These results were validated through placebo tests and other robustness checks. These findings confirm that reducing and/or eliminating taxes on smartphones improves smartphone penetration and contributes towards bridging the digital divide. The work can be extended by exploring VAT exemptions in other countries and reductions on other forms of taxes.

Link to Paper

Abstract:
Current economic and political factors, aligned to keep a geopolitical balance of power between the countries that produce telecom equipment – companies such as Ericsson, Nokia, Samsung, Mavenir, NEC and Fujitsu vs. Huawei and ZTE – are providing a challenging battle when it comes to emergent markets in developing countries. Worldwide there are wide arguments that Huawei and ZTE deliver better and cheaper equipment with usually lower rates of borrowing from the Chinese financial service industry. Furthermore, it is implied that other vendors are more expensive and if chosen the cost will hinder the expansion of infrastructure in rural areas.

This paper will present the result of the evaluation of data sourced from both regulators and industry in emerging markets on the choices made by telecom infrastructure providers, when to select or decline a vendor to be a trusted partner. For this purpose, we have reviewed both policy and industry actions in Latin America and the European Economic Area as both are representative of the challenges telecom infrastructure investment confront. Source documents revised in English, Spanish and Portuguese include reports and evaluations for partners section in those two regions.

Our analysis calls for a more careful appreciation of such statements. Instead of focusing just on the pricing of active equipment alone which makes a share of total cost of a radio access site or an end-to-end network, a more comprehensive approach to analysis is to compare the actual difference in costs, taking into consideration all factors, including the coverage, energy efficiency and quality network operations can provide. We also gathered evidence that the saving on buying infrastructure by the operator tends to improve their free cash flow that is not used to buy more equipment or improve coverage, but to increase the share price and dividends to shareholders.

The evidence collected points to some observations that require careful evaluation, take the case for example the comparison of CAPEX vs Sale and indicator of potential scale of impact on end-user prices. The more comprehensive approach in this paper takes into account what does it cost to build infrastructure, how much of that money goes into RAN equipment and how much goes to steel towers, power, and blue collared workers and other costs that are independent of the vendor chosen. There is also a need of determining if it is really the case that the operators in e.g. Latin America or Africa who do not use Chinese equipment have poorer network / coverage, and if it is the case that the operators using Chinese equipment have better networks.

The European Union has elaborated in the last few years policies and restrictions towards privacy and security breaches by Chinese telecom equipment and has successfully enforced on telecom operators’ requirements to include comprehensive estimates (E.g. Costa Rica).

Decision making when it comes to vendors require to make in-depth questions such as: What are the other hidden costs associated with choosing a non-trusted vendor like Huawei or ZTE? The distribution of such costs between the society and private telecom operators of infrastructure is of importance, to understand private companies’ incentives and risk incentives. Furthermore, the issue of trust is very relevant when it comes to choosing partners and short and medium term “savings” in infrastructure development in telecoms are not easy or cheap to rectify once the network is deployed to comply with regulation from countries or regions – such as the European Economic Area - that have objections to networks managed by non-trusted vendors. Hence the so-called benefits can end up taking a significant toll in terms of costs and loans to comply with those requirements.

It is our belief that our main contribution is to provide a framework where all these factors are taken into consideration both regulators and network operators will carefully consider present and future choices of vendors worldwide. Since the goal is to create, deploy and maintain networks that are cost viable, economically profitable, and trusted to provide the privacy and security citizens of democratic countries are entitled to, as these will be the foundation of the evolving digital economy in emerging markets .

Link to paper

Abstract:
I measure across- and within-household inequalities in the adoption of digital technologies. I leverage nationally representative household surveys in 98 low- and middle-income countries to build a new dataset covering 2,030,308 households surveyed between 2015-2023 – a dataset representative of 4.2 billion of the world’s population. Where feasible, I link household-level surveys with within-household women’s surveys. I then use this dataset to provide harmonised survey-based estimates of mobile phone ownership and use (for financial transactions), computer ownership and use, internet access at the household, and women’s use. I document three patterns: 1) Women’s phone ownership lags household ownership across the wealth distribution, with particularly stark gaps in South Asia. 2) Even when women own phones, their reported use of the internet (ever) or for financial transactions is lower and heterogeneous across low- and middle-income countries. 3) While mobile phones have wide proliferation, computers and internet access are often much poorer.

The Broadband Equity, Access, and Deployment (BEAD) Program in the U.S. has directed $42.45 billion in funding to accelerate the development and deployment of broadband infrastructure that meets specific service requirements such as minimum service speeds of 100 Mbps in the downstream and 20 Mbps in the upstream, peak hour latency below 100 ms, and network reliability of no more than 48 hours of outage time over any 365-day period. An often-overlooked element to the development of these requirements are the substantial policy implications they carry for determining preferred local access network technology options, preferred broadband service operators based upon their legacy technology positions, and the overall degree to which the nation’s households are unserved and underserved in terms of broadband access. Moreover, the performance requirements can influence, the design and evolution of the Internet architecture itself by influencing the selection or implementation of protocols and measurement metrics used by broadband network operators. To date in the U.S., Broadband Internet Access Service (BIAS) performance requirements have been developed in a number of different policy efforts such as the FCC’s annual Section 706 inquiries since the 1996 Telecom Act, part of the FCC’s universal service broadband funding programs since the National Broadband Plan (beginning with CAF Phase I and II, ACAM, and then RDOF), or adopted for broadband funding programs passed by Congress (such as Treasury’s Capital Project Fund and now the NTIA’s BEAD program).
This panel will investigate various policy concerns that arise from the lack of a consistent policy framework for defining BIAS performance requirements by addressing the following questions:

1. Prior Efforts to Characterize Performance-Related Policy Concerns.
2. How has Internet Performance Changed Over Time?
3. Increasing Importance of Latency in Internet Performance using Internet Engineering Task Force’s (IETF) Low Latency Low Loss Scalable Throughput (L4S) open standards.
4. Anticipated New Requirements from Emerging Internet Applications.
5. Putting It All Together: How Should Broadband Funding Be Linked to Internet Performance Requirements?

Link to paper

Abstract:
The focus of this paper is on an economics-based analysis and redesign of mitigation of volumetric Denial-of-Service (DDos) attacks utilizing the initial packets in connection-oriented protocols. These attacks have been documented for decades and their volume and impact have been growing enormously. The contributions of the paper are: (1) a novel evaluation framework based on all of efficacy, threat, overhead, and damage as experienced by applications; (2) a comparison of TCP with SYN Cookies and QUIC with Retries, the two accepted DDoS mitigation approaches in these two transport protocols; (3) evaluation of our alternative proposal to use SYN Proof-of-Work (SYN PoW) to address these volumetric attacks more effectively; and (4) an in-depth discussion of the economics of the various stakeholders in these scenarios. As demonstrated in this work, the SYN PoW type of approach not only moves much of the cost of mitigation onto the attackers, unlike current proposals, but also enables verification of validity of traffic to be handled anywhere in the network, rather than only at the end-points, giving network service providers an additional capability for reducing malicious traffic. A critical contribution is that this type of approach

Link to paper

Abstract:
Scholarly inquiry has yet to comprehensively explore the impact of Internet connectivity levels on dynamics related to online echo chambers, trust in online social networks (OSNs), and concern by users for algorithmic manipulation. To investigate the relationship between OSN users’ attitudes regarding online echo chambers, trust in OSN platforms, and concerns about algorithmic manipulation and Internet connectivity, we conducted a survey of more than 700 OSN users across diverse contexts. We collected responses from the United States, United Kingdom, India, and South Africa about attitudes and beliefs toward encountering discordant information in online settings. We integrate our survey response data with existing datasets from the Global System for Mobile Communications Association (GSMA) which indexes infrastructure, affordability, consumer readiness, and content & services to analyze mobile connectivity in a country. Our study reveals marked differences in users’ reported concerns for prevalence of OSN echo chambers, trust in OSNs as a source for news information, and concern for algorithmic manipulation depending on the level of Internet connectivity in various geographic regions. Our findings will aid global policymakers in tailoring policies addressing the aforementioned and related Internet phenomena to suit the specific context of Internet accessibility within the targeted country.

Link to paper

Abstract:
The Internet has become a critical basic infrastructure for society and the economy. As such, the resilience of the Internet is essential for the functioning and resilience of many sectors of our digital economies. In our increasingly digital society and economy, Internet is a key element of cyber-resilience. Understanding how to assess and promote Internet resilience are important policy challenges. In this paper, we discuss key aspects of the Internet design that differentiate it from other critical infrastructure and that are essential in considering its resiliency. We also explore how to achieve resiliency and provide a series of illustrative examples that reveal the challenges of achieving and measuring Internet resiliency, as well as mechanisms that contribute to it. Finally, we offer our thoughts on the policy challenges and approaches for promoting Internet resiliency.