NHH Brage (Norges Handelshøyskole)
Not a member yet
8813 research outputs found
Sort by
CBAM in the Aluminium Industry: Transforming Global Trade and Emissions
Get PDFThe European Union’s Carbon Border Adjustment Mechanism, introduced under the 'Fit for 55' package, seeks to address carbon leakage and foster international climate cooperation by imposing carbon-related tariffs on imported goods. This thesis focuses on the aluminium industry, a high-emission sector highly vulnerable to the economic ramifications of CBAM. Specifically, it explores the mechanism’s economic, environmental, and trade implications. By utilizing 2023 import data and a scenario-based modeling framework, this study assesses the implementation of CBAM and its impact on the aluminium industry.
The findings of this thesis indicate that the proposed CBAM framework, targeting direct emissions, is likely to decrease imports from high-carbon intensity producers such as China, India, and the UAE due to the increased cost of carbon-intensive imports. This reduction is estimated to contribute to a 13% decrease in global emissions from the aluminium industry. Expanding CBAM to include indirect emissions is projected to further enhance its effectiveness, leading to even greater reductions in imports from high emission intensity producers. These imports are expected to be replaced by suppliers with low-carbon intensity, doubling the overall reduction in global emissions. This highlights the considerable potential of an expanded CBAM framework in addressing carbon leakage and promoting sustainable trade.
The Norwegian aluminium industry holds significant potential to expand its market share in the EU by offsetting reduced imports from high-carbon intensity producers. By utilizing its full production capacity, Norway could supply up to 1.54 MT of low-carbon aluminium, strengthening its role in sustainable material transitions. However, critical flaws in the current CBAM framework, including the scrap loophole, exclusion of downstream products, and treatment of indirect emissions, combined with the phase-out of free allowances and CO2 compensation, pose significant risks to competitiveness. Without substantial revisions to CBAM after the transitional period, these challenges could erode Norway's ability to leverage its low-carbon advantage in the EU market.The European Union’s Carbon Border Adjustment Mechanism, introduced under the 'Fit for 55' package, seeks to address carbon leakage and foster international climate cooperation by imposing carbon-related tariffs on imported goods. This thesis focuses on the aluminium industry, a high-emission sector highly vulnerable to the economic ramifications of CBAM. Specifically, it explores the mechanism’s economic, environmental, and trade implications. By utilizing 2023 import data and a scenario-based modeling framework, this study assesses the implementation of CBAM and its impact on the aluminium industry.
The findings of this thesis indicate that the proposed CBAM framework, targeting direct emissions, is likely to decrease imports from high-carbon intensity producers such as China, India, and the UAE due to the increased cost of carbon-intensive imports. This reduction is estimated to contribute to a 13% decrease in global emissions from the aluminium industry. Expanding CBAM to include indirect emissions is projected to further enhance its effectiveness, leading to even greater reductions in imports from high emission intensity producers. These imports are expected to be replaced by suppliers with low-carbon intensity, doubling the overall reduction in global emissions. This highlights the considerable potential of an expanded CBAM framework in addressing carbon leakage and promoting sustainable trade.
The Norwegian aluminium industry holds significant potential to expand its market share in the EU by offsetting reduced imports from high-carbon intensity producers. By utilizing its full production capacity, Norway could supply up to 1.54 MT of low-carbon aluminium, strengthening its role in sustainable material transitions. However, critical flaws in the current CBAM framework, including the scrap loophole, exclusion of downstream products, and treatment of indirect emissions, combined with the phase-out of free allowances and CO2 compensation, pose significant risks to competitiveness. Without substantial revisions to CBAM after the transitional period, these challenges could erode Norway's ability to leverage its low-carbon advantage in the EU market
"Co-Cre-AI-tion"
Get PDFDet er over to år siden den generative AI-modellen Chat GPT ble tilgjengelig for allmennheten. Siden den gang har over 200 millioner mennesker verden over tatt i bruk verktøyet, som for mange i dag er en integrert del av hverdagen. Kunstig intelligens (AI) berører mange bransjer i arbeidslivet, inkludert den kreative bransjen. Vår masterutredning tar for seg hvordan ansatte i den kreative bransjen beskriver endringer i arbeidsprosesser som følge av samarbeidet med AI, omtalt som co-creation, samt hvilke forventninger de har til fremtidens samarbeid. Gjennom en kvalitativ casestudie har vi gjennomført 14 semi-strukturerte dybdeintervjuer i en utvalgt casebedrift, som utgjør grunnlaget for analysen og funnene i masterutredningen.
Studien belyser behovet for å forstå hvordan AI kan påvirke kreativitet og produktivitet, samt endre arbeidsprosesser i en bransje der menneskelig originalitet og intelligens står i sentrum. Funnene avdekker deriblant at informantene beskriver AI som et nyttig verktøy som kan bidra til effektivisering av tidkrevende og standardiserte oppgaver, og som dermed frigjør tid til mer verdiskapende arbeid som kreativ problemløsning. AI kan fungere som støtte til menneskelig intelligens, men informantene peker på viktigheten av menneskelig kreativitet og kritisk bruk for å sikre kvalitet og originalitet når AI tar en større rolle i samarbeidet. Informantene kan videre se for seg at fremtidige arbeidsprosesser er mer preget av fullverdig co-creation mellom mennesker og AI, men påpeker at verktøyene i dag ikke er modne nok til å muliggjøre dette for fullt. Funnene viser også en forventning om endringer i arbeidshierarkier, der etablerte strukturer kan bli utfordret i tiden som kommer grunnet kunnskapsutjevning mellom ansatte. Til slutt understrekes et behov for interne- og bransjespesifikke reguleringer for å sikre ansvarlig co-creation i fremtiden.
Begrensninger ved studien innebærer begrenset erfaring blant oss som forskere, samt tids- og ressursbegrensninger knyttet til datainnsamlingen. Likevel mener vi at studien gir verdifull innsikt når det gjelder hvordan AI kan forme arbeidslivet i den kreative bransjen, og dermed hvordan det bør legges til rette for å integrere AI på arbeidsplassen. Fremtidig forskning bør blant annet utforske de kausale sammenhengene mellom AI-samarbeid og endring i kreativitet og produktivitet. Videre burde fremtidig forskning kartlegge ytterligere rundt hvilke faktorer som må være på plass i en organisasjon for å sikre god utnyttelse og implementering av AI
The New Law of Corporate Bonds; Survival of the Greenest? Evidence of Greenium in the US and European Bond Markets
Get PDFClimate crisis and enhancing sustainability now play a crucial role in the global economy. The international community is actively collaborating to devise effective strategies to strengthen a sustainable and rational green transition. Governmental bodies, the corporate sector, and financial markets are pivotal in this transition. This thesis investigates one innovative, sustainable financial instrument: Green bonds, primarily corporate green bonds, issued in the US and European markets.
A sample of 12,541 corporate bonds issued between 2013 and 2023 is investigated to see whether green bonds achieve lower funding costs than conventional bonds, a phenomenon known as greenium.
The study employs four regressions and three propensity score matching analyses, with yield to maturity as the dependent and treatment variable, representing issuers' funding costs. The models control for issue characteristics, issuer attributes, and market conditions at issuance. The regressions explore variations with interaction terms and different explanatory variables. In the propensity score matching, green bonds are the treatment group and conventional bonds the control group, with both average treatment effect and average treatment effect on the treated reported.
The analyses findings provide support for the presence of a greenium. The results from the regressions indicate differences in funding costs estimated to range between -50.4 bps and -35.8 bps. Propensity score matching yields estimated values of -21.8 bps and -16.7 bps as the average effect of issuing a green bond. Further refinements lead to an average effect of issuing a green bond, among issuers of green bonds, of -40.7 bps.
Additionally, the results document that the longer maturity of a bond, the higher the funding cost for the issuer. It is also shown that an increased term spread results in lower financing costs, and that higher yields on one-year government bills result in higher funding costs for issuers of corporate bonds.Climate crisis and enhancing sustainability now play a crucial role in the global economy. The international community is actively collaborating to devise effective strategies to strengthen a sustainable and rational green transition. Governmental bodies, the corporate sector, and financial markets are pivotal in this transition. This thesis investigates one innovative, sustainable financial instrument: Green bonds, primarily corporate green bonds, issued in the US and European markets.
A sample of 12,541 corporate bonds issued between 2013 and 2023 is investigated to see whether green bonds achieve lower funding costs than conventional bonds, a phenomenon known as greenium.
The study employs four regressions and three propensity score matching analyses, with yield to maturity as the dependent and treatment variable, representing issuers' funding costs. The models control for issue characteristics, issuer attributes, and market conditions at issuance. The regressions explore variations with interaction terms and different explanatory variables. In the propensity score matching, green bonds are the treatment group and conventional bonds the control group, with both average treatment effect and average treatment effect on the treated reported.
The analyses findings provide support for the presence of a greenium. The results from the regressions indicate differences in funding costs estimated to range between -50.4 bps and -35.8 bps. Propensity score matching yields estimated values of -21.8 bps and -16.7 bps as the average effect of issuing a green bond. Further refinements lead to an average effect of issuing a green bond, among issuers of green bonds, of -40.7 bps.
Additionally, the results document that the longer maturity of a bond, the higher the funding cost for the issuer. It is also shown that an increased term spread results in lower financing costs, and that higher yields on one-year government bills result in higher funding costs for issuers of corporate bonds
Chunk Smarter, Retrieve Better: Enhancing LLMs in Finance : An Empirical Comparison of Chunking Techniques in Retrieval Augmented Generation for Financial Reports
Get PDFThis thesis investigates how Retrieval-Augmented Generation (RAG) improves the ability of Large Language Models (LLMs) to filter information from financial documents. For this task, we first develop NorwegianFinanceQA, a dataset containing 433 queries from the financial reports of 9 Norwegian companies, divided into text- and table-related queries. Next, we evaluate the retrieval accuracy and efficiency of RAG systems with different chunking techniques: character-based, recursive, and semantic splitting. Additionally, we propose a table-specific summarization approach. Our results suggest that table summaries achieve perfect accuracy for table queries while at the same time increasing efficiency. However, this improvement comes at the expense of text-query performance. Our findings highlight the importance of tailored chunking strategies when using LLMs and RAG systems for information retrieval in a financial context
Strategy and Digital Transformation: Exploring the Relationship Between Strategy and Digital Transformation in Established Firms
Get PDFThe purpose of this thesis is to investigate how strategy and digital transformation work together
to enhance firms' ability to compete in a digital landscape. Digital technology has
fundamentally altered market dynamics and product/service offerings, making a firm’s ability
to drive digital transformation a critical source of competitive advantage. However, not all
established firms have successfully capitalized on these opportunities. While some studies
emphasize the challenges of using strategy to drive digital transformation within established
firms, others point to the obstacles of effectively leveraging new digital resources resulting from
digital transformation to form strategies for growth and competitiveness. These questions are
still not settled, and more research is needed.
This thesis addresses these gaps with one conceptual and two empirical papers. The first
empirical paper examines the heterogeneity in firms’ digital strategy by introducing digital
orientation as a new antecedent and contrasting it with market orientation and entrepreneurial
orientation. Using textual analysis of annual reports from 73 Norwegian publicly traded
companies over ten years to operationalize the concepts, their relationships are analyzed
through a longitudinal structural equation model. In the short term, we find a circular
relationship whereby entrepreneurial orientation increases digital orientation, digital increases
market orientation, and market orientation increases entrepreneurial orientation. No long-term
relationship between the variables is observed. The second empirical paper hypothesizes that
incentives for digitalization vary by stage of digitalization. Using survey responses from 823
Norwegian firms during COVID-19, analyzed with ordered probit and multivariate probit
regression models, we find that digital leaders accelerated their digitalization efforts during the
pandemic, indicating a polarizing effect between leaders and laggards. The third, sole-authored,
conceptual paper addresses the paradox in optimal growth strategies for digital firms,
contrasting focus strategies through digital scaling with diversification through platform
envelopment and bundling. A new conceptual model suggests that optimal strategies are
contextual and determined by a firm’s relative level of digital transformation against
competitors.
In sum, this thesis contributes to strategic management literature, which is informed by
IS literature, providing new conceptual and empirical insights into how strategy and digital
transformation work together. It explains why some firms are more proactive in building
strategies for digital transformation, why some accelerate their digitalization faster than others,
and how they may optimize their digital growth strategies as a result
Implications of the resource rent tax on onshore wind
Get PDFTo minimize disruptions to the socially optimal allocation of resources, it is essential to design taxes that are as neutral as possible. Resource rent, a super-profit generated from exploiting scarce natural resources, can theoretically be taxed neutrally. Successfully designing a neutral tax would enable the collection of necessary taxes to fund the welfare state without influencing investment decisions.
This thesis examines how the resource rent tax impacts future investments in the onshore wind industry and the broader consequences of the tax. The analysis is based on realistic cash flow models and sensitivity analyses. Data sources include survey responses from six wind power companies and five renewable energy analysts, in-depth interviews with four wind power companies, a consultation with a specialist auditor from the Norwegian Tax Authority, and data published by NVE. Since the tax framework has yet to be finalized, the research considers two possible scenarios. The difference between these scenarios is whether the government will provide payouts for negative resource rent tax.
In the scenario without payouts, the resource rent tax reduces the IRR by 3% from 6.3% to 6.1%. The tax does not pose a financial barrier and should have a minimal effect on future investment activity from a financial perspective. In the scenario where the government provides payouts for negative resource rent tax, the tax increases the IRR by 17% from 6.3% to 7.4%. This is a significant increase and should contribute to higher investment activity.
The introduction of the resource rent tax has increased the perceived political risk associated with Norway and reduced its competitiveness as an investment destination. This has led to a decline in international ownership and potentially reduced the country’s ability to attract capital for new wind projects. Additionally, the tax has failed to improve local acceptance, a key bottleneck for onshore wind development. There is also a potential spillover effect as there are concerns that the resource rent tax could eventually be implemented for offshore wind. This uncertainty about future regulatory conditions can stop new investments in the offshore wind sector, as investors are unsure of the framework governing their investments. Combined, these consequences threaten to slow wind power development in Norway, potentially shifting the country's energy balance from a surplus to a deficit
Savings Banks or Super Banks? An Empirical Study on the Impact of Equity Certificates’ Unique Features on Pricing
Get PDFEquity Certificates (“ECs”) have been pivotal in providing Norwegian savings banks with
access to capital and maintaining their competitive market position. ECs are distinctively
characterized by their complex equity structure, downside protection, and limited voting
rights. Despite ECs’ significance, their pricing dynamics have been subject to meager academic
research, partly due to being confined to the small Norwegian market. This motivated us to
conduct a preliminary empirical study on the impact of ECs’ unique features on pricing.
Specifically, we hypothesize that (1) the market values the downside protection in ECs, and
(2) their unique features contribute to pricing discrepancies compared to shares.
We investigate our hypotheses through panel data regressions utilizing the Price-to-Book
(“P/B”) ratio as a dependent variable. For all models, we utilize a self-constructed novel
dataset of all listed Norwegian savings banks with 1764 unique bank-quarter observations,
from 2005-Q1 to 2023-Q4. This research is an initial step toward building a solid foundation
rooted in empirical analysis on the subject of EC pricing dynamics.
We find a positive relationship between the downside protection of ECs and the P/B ratio.
However, we find inconclusive evidence that this positive effect is greater during periods of
financial instability, contrary to what financial theory would suggest. Our models’ inability
to yield conclusive results during crises may be attributed to ambiguous interpretations of the
downside protection proxies in such periods. Surprisingly, being an EC bank, as opposed to a
share bank, seems to negatively influence pricing. This implies that ECs’ downside protection
is overshadowed by features such as instrument complexity and limited voting rights. Finally,
our quasi-experiment yields inconclusive results on the effect of converting ECs to shares.
Overall, the thesis concludes that the market understands the unique equity structure of ECs,
at least to some extent, and prices them accordingly. Beyond the pricing dynamics, our
research raises questions about whether the capital structure serves its intended purpose.
Thus, the findings of our thesis may offer meaningful insights to key stakeholders.nhhma
Country-of-Origin Effects in B2B Markets
Get PDFThis study investigates the country-of-origin (COO) effect on organizational buyers. Five hypotheses were developed to test how Norwegian COO information influences B2B buyers’ evaluations, purchase intentions, and perceptions, with a focus on factors such as presentation method, complexity, familiarity and subjective norm. The effects of these variables were assessed using a three-group experimental design (control group, COO label, and COO added benefit link) in the context of two sectors (grocery/retail/wholesale and medical/healthcare), where respondents viewed a product advertisement followed by a questionnaire.
The results indicate that exposure to COO information enhances organizational buyers' evaluations of Norwegian products, although this effect appears to be product dependent. Adding a benefit link that highlights a specific Norwegian advantage did not yield additional positive effects in this study; in some cases, it even showed a negative impact. The influence of product complexity was found to be non-significant, potentially due to the relatively low technical complexity of the tested products (facemasks and salmon). Product familiarity influenced the use of COO information, but the effects were mixed, showing both positive and negative outcomes. Finally, our findings on the mediating effect of subjective norm were inconclusive, likely due to the theoretical challenges of accurately measuring subjective norms through questionnaires.
This study offers new insights into the relatively modest research area of COO effects in B2B markets. Specifically, it provides evidence of favorable COO effects for certain Norwegian products, such as facemasks. Furthermore, the study suggests that these enhanced evaluations may arise from factors like the higher potential for perceived quality improvement and the perception that such products are typically associated with newly industrialized countries. These insights could prove valuable for Norwegian companies determining whether to implement the “Made in Norway” label as a marketing tool
KI i kommunal sektor
Get PDFDen raske utviklingen innen kunstig intelligens (KI) skaper muligheter for å forbedre effektivitet og kvalitet i offentlig sektor, men medfører også betydelige utfordringer. Med bakgrunn i dette søker vi i denne avhandlingen å besvare forskningsspørsmålet: "Hvordan forbereder norske kommuner seg på å ta i bruk kunstig intelligens?".
Bergen kommune var et naturlig valg som case da kommunen har ambisjoner om å være en ledende aktør innen digitalisering i offentlig sektor. Studien undersøker i hovedsak kommunens forberedelser, men kaster også lys over muligheter og medførende barrierer, og fremtidsstrategier for KI-implementering.
Studien har en kvalitativ tilnærming og benytter semistrukturerte intervjuer og dokumentstudier for å samle data. Informantene inkluderer ansatte i Bergen kommune med innsikt i digitaliseringsprosesser og kunstig intelligens. Analysen fokuserer på fem kategorier for KI-beredskap, basert på rammeverket til Jöhnk, Weißert og Wyrtki (2020): strategisk tilknytning, ressurser, kunnskap, kultur og data.
Hovedfunnene viser at Bergen kommune har gjort fremskritt i å forberede seg på KI, spesielt gjennom dataforvaltning. Kommunen har også tatt initiativ til innovasjonsarbeid og samarbeid med eksterne aktører. Samtidig identifiseres flere barrierer, inkludert begrensede økonomiske og menneskelige ressurser, samt organisatorisk treghet. Ansattes kompetanse og aksept for teknologien varierer, og det er behov for ytterligere opplæring for å sikre en vellykket implementering. Følgende har studien re-konseptualisert rammeverket for KI-beredskap til å inkludere tre kontekstspesifikke faktorer, nemlig eksternt samarbeid, etablering av testmiljø og strukturering av data.
Studien gir innsikt i hvordan norske kommuner kan styrke sin KI-beredskap ved å adressere organisatoriske, teknologiske og kulturelle forhold. Funnene har overføringsverdi til andre kommuner, særlig de som står overfor lignende ressurser og strukturelle begrensninger.The rapid development of artificial intelligence (AI) creates opportunities to improve efficiency and quality in the public sector, but also poses significant challenges. With this in mind, this thesis seeks to answer the research question: “How are Norwegian municipalities preparing to adopt artificial intelligence?”.
The municipality of Bergen was a natural choice as a case study as the municipality has ambitions to be a leading player in digitalization in the public sector. The study mainly examines the municipality's preparations, but also sheds light on opportunities and associated barriers, and future strategies for AI implementation.
The study has a qualitative approach and uses semi-structured interviews and document studies to collect data. The informants include employees in Bergen municipality with insight into digitalization processes and artificial intelligence. The analysis focuses on five categories of AI readiness, based on the framework of Jöhnk, Weißert and Wyrtki (2020): strategic fit, resources, knowledge, culture and data.
The main findings show that Bergen Municipality has made progress in preparing for AI, especially through data management. The municipality has also initiated innovation work and collaboration with external actors. At the same time, several barriers are identified, including limited financial and human resources, as well as organizational inertia. Employee competence and acceptance of the technology varies, and further training is needed to ensure successful implementation. Consequently, the study has re-conceptualized the AI readiness framework to include three context-specific factors, namely external collaboration, establishing a test environment and structuring data.
The study provides insight into how Norwegian municipalities can strengthen their AI readiness by addressing organizational, technological and cultural factors. The findings have transferable value to other municipalities, especially those facing similar resource and structural constraints
Does Norges Bank’s Crystal Ball Work?
Get PDFWith this thesis we conduct a novel examination of Norges Bank’s policy rate forecasts. We find that the quality of Norges Bank’s forecasts drops severely already after two quarters. Our analysis suggests that a bias towards overpredicting the policy rate leads to forecasts losing efficiency from the third quarter. We apply methodologies specifically developed for the purpose of evaluating forecasts to check for a bias toward underestimating the momentum of interest rates. Our findings show that even when forecasts are made and realized within the same interest rate cycle, forecasts regularly underestimate the momentum of interest rates. This leads to forecasts made in an upturn being too low, and forecasts made in a downturn being too high. This finding is completely in line with other studies into forecasting of a cyclical variable, but has as far as we know never been uncovered for Norges Bank previously. To Norges Bank’s credit, we find that they are more accurate forecasters than markets for the first five quarters, but the difference is only statistically significant for the first two quarters. We also find that incorporation of Norges Bank’s forecasts into market forecasts improves market accuracy for all forecasts horizon, although only in a significant way for a few forecast horizons