The International Accounting Standard 17 (IAS 17) was introduced in 1892. It required both the legal owner of the asset (lessor) and the user of the asset (lessee) to make a distinguishment between an operating or a finance lease (IASB 2001). However, IAS 17 did not clearly define how to classify a lease as financial, resulting in most leases being classified as operating. This difference in classification was clear, but the difference between the accounting for finance and operating lease was material and significant. For example, for an operating lease, it was not required to raise a lease asset and a liability on the balance sheet, resulting in the accounting of operating leases being off-balance (Goodacre 2003).

In 2016 the International Accounting Standards Board (IASB) issued a new standard for leasing, and as of the financial year 2019, companies were obliged to report under International Financial Reporting Standards (IFRS) 16. To ensure that leases were no longer off-balance sheet, IFRS 16 requires the companies to treat all leases as finance leases. Companies are now required to recognise a Right of Use (ROU) asset with a corresponding liability on their balance sheet for all their leases (Milton and Genevieve 2019). The IASB introduced IFRS 16 to improve reporting quality and give a more faithful representation of financial statements (KPMG 2021). Even though IFRS 16 does give a more faithful representation of the financial statements, the financial ratios have changed drastically. As mentioned, the airline industry had significant changes in its financial statements due to IFRS 16. If you look at Air France, the national airline of France, their EBITDA has increased by 29.2% (2017: 3.264 million, 2018: 4.217 million). 2017 was chosen because Air France was an early adopter of IFRS 16 and had been reporting under IFRS 16 since 2018 (Air France-KLM Group 2017; 2018). Air France-KLM Group (2018) found that the implementation of IFRS 16 had implications for the opening and ending balance “especially in the Right-of-Use assets and the lease debts” (p.17). Besides EBITDA, other ratios have also changed drastically, and this is a concern to many organisations because financial ratios measure an entity’s performance and creditworthiness. This change in ratios can significantly impact the decision potential investors might make (Altamuro et al. 2014).

Bank loans include, besides a price term, also a non-price term. Items like loan size, maturity or restrictive covenants are terms that lenders use to consider whether or not to give out a loan and what the terms of the loan are. Lenders also use these terms in loan contracts to extenuate potential agency conflicts and information problems. Lenders control their exposure to risk by reducing loan size and shortening loan maturity (Kim et al. 2011). Chen et al. (2015) found that companies adopting IFRS are given longer maturities and loan sizes. Kim et al. (2011) and Ball et al. (2015) also found evidence that adopting IFRS reduced the uncertainty of information asymmetry between lenders and borrowers. IFRS 16 is designed to give an even more faithful representation and reduces this information asymmetry probably even more. This could imply that companies reporting under IFRS 16 are given larger loan sizes and longer maturities since there is less information asymmetry and more transparency.

Another expectation that is derived from Kim et al. (2011) is that the cost of external financing is reduced by high-quality information. Lenders view the adoption of IFRS as a commitment to better reporting strategies and are, therefore, more eager to offer favourable contracts to lessees. However, Milton and Genevieve (2019) concluded that IFRS 16 users are generally considered less creditworthy and less inclined to get loans with a favourable borrowing rate. However, as research from Altamuro et al. (2014) showed, lenders generally do not base their decisions on the financial statements or ratios alone and therefore, this research predicts the following relationship regarding (non-)price terms of loan contracts:

H1: Companies reporting under IFRS 16 are given larger loan sizes

H2: Companies reporting under IFRS 16 are given longer loan maturities

H3: Companies reporting under IFRS 16 are given lower borrowing rates

People with diverse access to financial resources have various capacities for acquiring and processing information when knowledge about financial assets is costly to analyse. Sophisticated lenders generally have more information available and are expected not to experience differences from the new accounting standard for IFRS 16. Determining when lenders are classified as sophisticated or unsophisticated is also essential. Research from Che (2018) showed that informed investors are usually institutional investors. This could lead to a conclusion that institutional investors are sophisticated lenders since research from Kacperczyk (2014) showed that sophisticated lenders are usually better informed than unsophisticated lenders. Examples of institutional and thus sophisticated lenders are banks, trust companies, saving institutions, insurance companies or pension funds (Law Insider n.d.).

The research from Du and Palia (2018) stated that banks generally give out more long-term debt. This leads to the assumption that banks are sophisticated lenders and give out long-term debt. Trade creditors, however, give out more ‘public’ debt because they have an information asymmetry problem. Therefore, Bontempi et al. (2020) found that trade creditors give out short-term debt, resulting in the assumption that unsophisticated lenders, with more information asymmetry, resemble trade creditors more and give out short-term debt.

Sophisticated lenders can request additional information from the lessee, such as a separate P&L of the balance sheet, which eliminates the differences that occurred due to a change in accounting standards. This could mean that they can base their decisions regarding the loans on this additional, usually voluntarily, disclosed information. Unsophisticated lenders have less information and base their decisions on what is available publicly (Bandara and Falta 2021; Barber et al. 2009; Grinblatt and Keloharju 2000). In this research, sophisticated lenders are classified as banks and unsophisticated lenders are classified as trade creditors. The change in accounting standard is expected to be less impactful for sophisticated lenders than for unsophisticated lenders. Therefore, in this research, the following relationship is predicted and tested:

H4: IFRS 16 has a reinforcing effect on loan contracting for sophisticated lenders

Y₁ = β0 + β1 IFRS_i × AFTER_t + β2 IFRS_i + β3 AFTER_t + β4 ROA_i,t + β5 LEV_i,t + β6 SIZE_i,t + β7 Curr_ratio_i,t + β8 Big4_i,t + ƒ_i + δ_i,t + ε_i,t

Y₂ = β0 + β1 IFRS_i × AFTER_t × Sop_lender_i + β2 IFRS_i × AFTER_t + β3 IFRS_i × Sop_lender_i + β4 Sop_lender_i × AFTER_t + β5 IFRS_i + β6 AFTER_t + β7 Sop_lender_i + β8 ROA_i,t + β9 LEV_i,t + β10 SIZE_i,t + β11 Curr_ratio_i,t + β12 Big4_i,t + ƒ_i + δ_i,t + ε_i,t

In Y_1, hypotheses 1, 2 and 3 are tested using the difference-in-difference method. Y is equal to the dependent variables Maturity, LoanSize and Borr_rate tested in the first three hypotheses. The coefficient on IFRS_i measures the effect of the dependent variable (Y) being a company reporting under IFRS. The coefficient on AFTER_t measures whether there are changes in Y before 2019 and after. The most important effect to measure is the IFRS_i × AFTER_t, which measures the effect of the change in 2019 for companies reporting under IFRS on Y.

In Y_2, hypothesis 4 is tested using a difference-in-difference method with a triple interaction. Y is equal to the dependent variables Maturity, LoanSize and Borr_rate which are also tested in the first three hypotheses. However, now the variable Sop_lender_i is added as a moderating variable, creating the triple interaction of IFRS_i × AFTER_t × Sop_lender_i. With this triple interaction, the effect of the change in lease standard in 2019 for companies reporting under IFRS on Y is tested, and the effect the Sop_lender has on that is added as a moderating variable.

Both regression equations include firm-fixed effects (ƒ_i,t) , which account for unidentified firm-level time-invariant heterogeneity (Vanhaverbeke et al. 2022).

To test H1 and H2, two nonprice terms of loan contracts will be used based on the research of Kim et al. (2011) and Chen et al. (2015). The first variable is the logarithm of the amount of a loan, calculated as the short-term debt plus the long-term debt, and is defined as LoanSize. The second variable used to test this hypothesis is Maturity. Maturity is the difference in months between the original loan date and the maturity date.

To test H3, the variable Borr_rate is used. The Borr_rate is calculated as the average interest rate lenders charge throughout a year with multiple loans (Degryse et al. 2016). Borr_rate reflects a loan’s perceived level of risk, also called credit risk. This rate is set up by lenders based on knowledge of the business’s nature and performance (Kim et al. 2011).

To test H4, all variables from H1, H2 and H3 are used, but the type of lender is added as a moderating variable. When looking at the type of lender, you can look at sophisticated and unsophisticated lenders. Bontempi et al. (2020) found that banks give out long-term debt, and trade creditors give out short-term debt. Therefore, the short-term and long-term debt to total-debt ratio is used to research the difference between sophisticated and unsophisticated lenders.

Besides these hypothesis-specific variables, some control variables will be used. These control variables were established similarly to Kim et al. (2011). The first control variable is Return on Assets (ROA), defined as the net income divided by total assets. The second control variable is Leverage (LEV) which represents the leverage calculated as the total debt divided by total assets. These control variables have been added to control for credit quality based on the research of Kim et al. (2011). Another control variable added is the firm size, calculated as the natural logarithm of total assets (Financial research data services 2021; Byard et al. 2011; Kim et al. 2011). Based on Chen et al. (2015), the control variable Curr_ratio is added and is calculated as the current assets divided by the current liabilities. To control for cross-firm differences in the environment, the control variable Big4 will be included, which equals one for firms audited by Big4 firms and zero otherwise.

Borr_rate and Maturity can be calculated and are available in the CAPITAL IQ – Capital structure debt database. LoanSize and the distinction between sophisticated lenders and unsophisticated lenders can be calculated using the Compustat Global – Fundamental annual database. The databases will be merged using the GVKEY, a company-specific code available in both databases. The control group will be divided from the treated group by a variable available in the Compustat Global database and states the accounting standard used by that company. A dummy variable can then be created where one is for companies using IFRS and zero for companies using US GAAP. All variables are displayed in USD for convenience since most variables are already in USD when extracted from the database.

A Difference-in-Difference analysis will be conducted to test the hypothesis outlined in this research. The difference-in-difference analysis can be used to research if and how lenders’ attitudes have changed after the introduction of IFRS 16. With a difference-in-difference analysis, it is easy to compare a treated group with an untreated group and get the effect of an event on the groups (Huntington-Klein 2022; Columbia public health 2023). In this research, the treated group are IFRS users since they changed their accounting. Since IFRS 16 is mandatory for all firms reporting under IFRS, it weakens potential self-selection or endogeneity concerns. Besides IFRS being mandatory for many firms, it is also broadly adopted, which ensures that there is a sample of companies worldwide, allowing a cross-sectional analysis (Chen et al. 2015). The intervention point will be the date IFRS users were obliged to report under the new leasing standard, IFRS 16. The time horizon for this research is three years pre- and post-intervention point, meaning data will be used from 1 January 2016 until 31 December 2021.

The untreated group, or control group, are US GAAP users. Since US GAAP users did not change their lease accounting, they mitigate potential confounding factors, such as economic conditions, and are an excellent control group in this research. The eventual result from the difference-in-difference analysis will tell the effect of the new lease standard by looking at the change in the control group and, thus, how much change is expected in the IFRS group if there had not been a new standard. The additional change post-implementation is then contributed to IFRS 16 and will tell us something about the effect of IFRS 16 on the attitude of sophisticated and unsophisticated lenders towards loan contracting (Huntington-Klein 2022).

In line with the research of Chen et al. (2015), all missing values for the accounting variables are removed. In line with the research of Altamuro et al. (2014), I remove financial firms (SIC codes between 6000 and 6999) because the credit assessment for industrial firms differs from the credit assessment for financial firms. In line with the research from Kim et al. (2011), I also remove firms in the public sector (SIC codes between 9100 and 9999). Besides that, firm-year observations with a negative maturity are removed since they have already elapsed, or the data might be incorrect and unreliable. Another essential aspect to consider is that every firm has observations for all six years of the time horizon. Therefore, firms that only have an observation in some of the six years of the time horizon have been removed. This also removes firms that have observations in 2015 or 2022.

To distinguish between IFRS and US GAAP, all rows that do not report using either US GAAP or IFRS are removed. Firms listed on the US stock exchange must report under US GAAP (Financial Accounting Foundation, n.d.), so I looked at which stock exchange the firms were listed for the Compustat databases and added them to the control group if they were listed on the US stock exchange. This left 23.733 firm-year observations of IFRS users and 8.600 firm-year observations of US GAAP users. Lastly, I winsorize all variables at 1 and 99% to normalize the sample.

After the data is collected and cleaned, the data is analysed. Table 2 gives the descriptive statistics for the full sample, the sample with solely IFRS users and the sample with solely US GAAP users. One crucial remark can be made about descriptive statistics. The difference mean of Maturity in the US GAAP group is greater by almost 40 months compared to the IFRS users group and the entire sample group.

The results of the regressions show some interesting findings. As predicted in hypothesis 1, implementing IFRS 16 could lead to larger loan sizes. However, these results were insignificant and not sufficient to support the first hypothesis and therefore the null hypothesis is accepted. In the second hypothesis I looked at the maturity which was significant but not in the expected direction, implying that IFRS users would have lower maturities after 2019. This is different from what has previously been found by Chen et al. (2015) and Kim et al. (2011). Both articles found that firms would get higher loan sizes and longer maturities after adopting IFRS. This is not the case when IFRS 16 is adopted, so the second hypothesis is rejected.

The third hypothesis expects that firms are subject to a lower borrowing rate when reporting under IFRS after the implementation of IFRS 16. Altamuro et al. (2014) and Kim et al. (2011) found that lenders view the adoption of IFRS as a commitment to better reporting strategies and are, therefore, more eager to offer favourable contracts to lessees. Lower borrowing rates are considered favourable, so it is expected that the implementation of IFRS 16 would lead to lower borrowing rates. IFRS users are charged lower borrowing rates after the implementation of IFRS 16.

For the fourth hypothesis, the moderating variable Sop_lender is added. The sophisticated lender is determined as banks since they have more information available and can base their decisions on additionally disclosed information (Bandara and Falta 2021; Barber et al. 2009). Therefore it is expected that IFRS 16 has a reinforcing effect on loan contracting. Based on the findings, the maturities are longer, and the borrowing rate is lower. This hypothesis only holds for these two variables because the loan size was reduced by sophisticated lenders. Since that result is not significant, the fourth hypothesis is still accepted.

One of the main objectives of the IASB when IFRS 16 was drawn up, was that it would give a more faithful representation of a company’s financials. With that in mind, it was expected that there would be larger loan sizes, longer maturities and lower borrowing rates. The borrowing rate was lower after the implementation of IFRS 16 for firms reporting under IFRS. Companies subject to lower borrowing rates can potentially reduce their borrowing costs, thereby improving their financial performance and access to capital in the future. The loan size was larger but insignificant, so no conclusion can be drawn on that aspect. Lastly, for maturity, it turned out that they became shorter after the adoption of IFRS 16. This might affect companies cash flow planning but most of all their financial flexibility. Companies must adapt their financial strategies by considering the potential implications this has on their operations. This was not expected based on the literature research and is different from the thought that a more faithful and transparent reporting rule would give more certainty to lenders and, therefore, would lengthen the term of a loan to spread risks.

What must be taken into account is that all dependent variables are related to each other and decisions about these loan terms are taken simultaneously at the time of loan origination. Therefore you could look at all these components separate, but in the end one overall decision must be made. In this research the conclusion for loan contracting as a whole would be that IFRS 16 has a positive effect on loan contracting. This means that companies got more favourable contracts with lower borrowing rates, larger loan sizes and shorter maturities. Overall, it can be said that companies are given better contract terms then before the implementation of IFRS 16.

When the moderating variable sophisticated lender is added, the expectation is that IFRS 16 would have a reinforcing effect for sophisticated lenders, meaning even longer loan sizes, longer maturities and lower borrowing rates. What was immediately apparent was that in terms of loan size, the sophisticated lender did not give larger loan sizes but even lower loan amounts than unsophisticated lenders. For the borrowing rate, this was not the case. As expected, the borrowing rate that sophisticated lenders charged was lower than that of unsophisticated lenders; therefore, IFRS 16 has a reinforcing effect on sophisticated lenders. Lastly, I looked at the attitude the sophisticated lender has towards the maturity of a loan. With the regular regression, it turned out that IFRS 16 would lead to shorter maturities. However, when you look at the attitude of the sophisticated lenders, maturities were longer after the implementation of IFRS 16 for companies reporting under IFRS. Therefore it can also be concluded that IFRS 16 has a reinforcing effect on the attitude of sophisticated lenders when you look at borrowing rates and maturities.

The expectation was that lenders would have more trust in companies since their financial reports are showing their lease liabilities completely and therefore give a better overview of a company’s financial situation which gives lenders a more faithful and complete picture. Unfortunately this was not directly shown from the results of this research, but even though the results did not indicate directly that IFRS 16 gives a more faithful representation resulting in more favourable loan contracting, there is still reason enough to believe that IFRS 16 has improved the reporting quality, since the financial statements now give a more complete overview of all the liabilities a company might have, and thus gives a more faithful representation of a company’s debt position.

This research also has some limitations. The main limitation was the reorganisation of the Thomson Reuters Dealscan database. This database generally contains more detailed information on loan contracting, such as lender types and restrictive contract covenants. However, this database was no longer helpful since many company-specific identifiers were mixed up or unavailable in the other databases used in this research. Another limitation of this research is that some firms are early adopters of IFRS 16. An early adopter reports under IFRS 16 before it is mandatory to use the standard. Unfortunately, this is not specified in the Compustat database; therefore, I could not make that distinction. The third limitation of this research is that COVID-19 might bias the results. Since COVID-19 only happened in 2020, it only affects the period after it was mandatory to report using IFRS 16. Besides that, COVID-19 is a worldwide pandemic. Therefore there is no control group available that did not undergo the effects of this pandemic, and therefore, I cannot control for this effect.