Loss Reserves Research Articles

Loss Reserving Prediction Error with Special Reference to a Tweedie Sub-Family

The paper is concerned with model distribution error in loss reserving, i.e. the error in the loss reserve that arises from the wrong choice of distribution of observations. This is considered within the Tweedie family of distributions, examining the prediction error that occurs when one value of the Tweedie dispersion parameter p is correct, but a different one assumed in the modelling a claim array. The study is carried out under the condition that the ratio of mean to variance is constant across all cells of the array, a condition that is found commonly compatible with real data sets. The main result of the paper is that, when cells have largish coefficients of variation under the constant mean-to-variance condition, and the array can be modelled with a GLM, the MSEP of the loss reserve is relatively insensitive to the value of p. This implies that model distribution error can often be dismissed as small in many situations where MSEP of loss reserve is the measure of interest.

Investigation of the Pillars of Sustainability Risk Management as an Extension of Enterprise Risk Management on Palestinian Insurance Firms’ Profitability

This research paper examined the simultaneous relationship between sustainability risk management (SRM) as an extension of Enterprise Risk Management (ERM) and Palestinian insurance firms’ profitability, for the period spanning 2007Q1 to 2018Q4, by applying the panel dynamic (Generalized Method of Moments) GMM model. The literature was expanded by providing a comprehensive understanding of determining the pillars of ERM with the use of the factor analysis principle component method. The findings revealed that the firm’s profitability positively corresponded to ERM1 implementation, which represents “management efficiency”. In contrast, it shows negative correspondence to ERM2 implementation, which represents “control and ownership”. Furthermore, there were slightly negative signs from managing the use of leverage and they were conservative in terms of loss reserves. The challenges of firms’ profitability have negatively corresponded to emerging sustainability risks, such as political stability, that cause premiums written to show weak signs of excessive choice of risk or prices that are not met carefully. Interestingly, there is a positive relationship in the interaction between ERM2 implementation during the crisis period on insurance firms’ profitability. There is a robust causal relationship from ERM to the profitability (either positive or negative). The reverse causality is also significant but to a lesser extent. Thus, the study recommends alignment more coherent with the implication of ERM as holistic risk according to the market characteristic towards the environmental perils leads to sustainable development and its segments to maintain the longer term of survival in the firms’ performance.

Determinants of banks’ Nerlovian economic efficiency: a DEA-bootstrap approach

ABSTRACT This study aims to evaluate the Nerlovian economic efficiency of Taiwanese commercial banks and its determinants by assuming the presence of an imperfectly competitive market using a two-stage estimation procedure: Nerlovian economic inefficiency and its components’ price, technical and allocative efficiencies computed and decomposed in the first stage, which are regressed on the explanatory variables with a bootstrapped truncated approach in the second stage. The estimation results show that in the first-stage analysis, the Nerlovian economic inefficiency of banks is primarily due to allocative inefficiency, and indicate the existence of price inefficiency in Taiwan. In the second-stage analysis, the results confirm that both the years in operation of the bank and the ratio of credit loans are the main determinants of banking profit efficiency. In addition, this study not only shows that publicly owned banks contribute to better price efficiency but also proves that loan loss reserve to total assets is negatively associated with technical efficiency. The equity ratio exerts an insignificant favourable impact on allocative efficiency. The findings of this research are essential for bank managers in Taiwan.

Stochastic loss reserving: A new perspective from a Dirichlet model

AbstractForecasting the outstanding claim liabilities to set adequate reserves is critical for a nonlife insurer's solvency. Chain–Ladder and Bornhuetter–Ferguson are two prominent actuarial approaches used for this task. The selection between the two approaches is often ad hoc due to different underlying assumptions. We introduce a Dirichlet model that provides a common statistical framework for the two approaches, with some appealing properties. Depending on the type of information available, the model inference naturally leads to either Chain–Ladder or Bornhuetter–Ferguson prediction. Using claims data on Worker's compensation insurance from several U.S. insurers, we discuss both frequentist and Bayesian inference.

Reducing Medical Malpractice Loss Reserve Volatility Through Tort Reform

This study examines how tort reform affects uncertainty in insurance markets by testing whether noneconomic damage caps influence reserving volatility for medical malpractice insurers. Using a panel of insurers from 1986 to 2009, we estimate the determinants of loss reserve error volatility and focus on how this volatility is influenced by the percent of premiums an insurer writes that are subject to noneconomic damage caps. We find empirical evidence that tort reform reduces reserve volatility over the subsequent 3- and 5-year periods, consistent with tort reform improving insurers’ loss forecasting ability. Our findings address outcomes of tort reform that are prominent concerns of legislators, regulators, and policyholders. This article contributes both to the literature examining the insurance market effects of tort reform and to the literature examining loss reserving practices.

Government Ability, Bank-Specific Factors and Profitability: An Insight from Banking Sector of Vietnam

To the best of our knowledge, a very few studies have focused on the effects of government ability on bank performance in developing and emerging countries. The aim of this work is to study the impact of government ability, bank-specific factors on profitability in the banking system in Vietnam. Using a panel data analysis in the period over 2014-2018, the study analyzes based on the methods of fixed, random effects, and pooled ordinary least squares. Data were collected from Vietnam’s Stock Exchange, General Statistics Office, and Worldwide Governance Indicators. Our results demonstrate that government ability has negatively affected bank efficiency while economic growth will not affect bank efficiency. In addition, the prime bank-specific factors that can significantly impact on bank efficiency are non-performing loan, loan-to-deposit ratio, loan loss reserves. A bank with a higher loan-to-deposit ratio can positively impact on the probability of a bank. In contrast to the risk, a bank with a greater risk as well as a higher level in non-performing loan in operation will negatively impact on its efficiency.

On bootstrap estimators of some prediction accuracy measures of loss reserves in a non-life insurance company

This article considers the hierarchical generalized linear model (HGLM) for loss reserving in a non-life insurance. In the current insurance practice, insurance companies use generalized linear models (GLM) for prediction of the total loss reserve. This model, however, requires the assumption of independence of random variables occurring in the loss triangle, which may not always be fulfilled. The remedy to this problem is to introduce random effects to the GLM allowing modeling dependence inside the loss triangle. A limitation in the use of HGLM to predict the total loss reserve is the fact that the error of prediction is expressed by a complex analytical formula. An alternative to the analytical approach is to use the bootstrap technique. The main purpose of this article is to propose the full residual and parametric bootstrap procedures for the estimation of three types of prediction accuracy measures. The first is a classic root mean squared error of prediction (RMSE). The second is a quantile of the absolute prediction error (QAPE). The third one, we propose, is a quantile of a mixture of absolute prediction errors. Stochastic properties of the estimators of the accuracy measures are studied in two Monte Carlo simulation experiments. Bootstrap procedures and Monte Carlo simulation are implemented in R program.

Current expected credit loss procyclicality: it depends on the model

The new guidelines for loan loss reserves, current expected credit loss (CECL), were initially proposed so that lenders’ loss reserves would be forward-looking. Some recent studies have suggested that CECL could be procyclical, meaning that loss reserves would peak at the peak of a crisis. Although it is better than seeing failure only after it has happened, being required to raise liquidity at the peak of a crisis could still fail to save the lender from collapse, and it may even facilitate the latter. However, previous procyclicality studies have explained all losses using macroeconomic factors, ignoring the changes in credit risk and other portfolio drivers that preceded the recession. The current work looks at a wide range of models to test the degree to which CECL is procyclical for different types of model. The tests were also run using real historical macroeconomic scenarios, flat scenarios or mean-reverting scenarios. All tests were conducted on publicly available data from Fannie Mae and Freddie Mac using publicly disclosed models. Our study found that CECL lifetime losses timates were only marginally sensitive to the quality of the economic scenario but changed dramatically with different modeling techniques. Some methods predicted increased loss reserve requirements as early as 2006, while others only saw the recession as it happened or even afterward. Therefore, procyclicality under CECL will be strongly influenced by the choices of the lender.

Policy uncertainty and loan loss provisions in the banking industry

Policy uncertainty is an increasingly important issue in many economies. Extensive evidence indicates that higher policy uncertainty is associated with future negative macroeconomic and microeconomic conditions. In this paper, we examine how policy uncertainty affects banks’ accruals for loan losses. Consistent with banks signaling more expected loan losses, we document that in times of higher policy uncertainty, banks make more loan loss provisions. This positive association is more pronounced for banks that were previously less prudent in their risk-taking and loan loss reserving, indicating that less prudent banks are harmed more by loan losses in difficult times. We also show that higher attention paid to a banks’ financial reporting strengthens the role of loan loss provisions as a signal of expected loan losses. Overall, our paper offers insight into how, in the face of policy uncertainty, banks convey information about their loan portfolios to their stakeholders.

The determinant of capital adequacy ratio: empirical evidence from Vietnamese banks (a panel data analysis)

The purpose of this study is to investigate the determinant of Vietnamese banks' capital adequacy ratio (CAR) by internal banking factors. Secondary data is collected from banks' annual reports in the period of 2009–2015. FGLS method and panel data are used to examine a regression model with CAR is the dependent variable and five independent variables: bank size (SIZE), loans (LOA), loan loss reserve (LLR), liquidity (LIQ), profitability (ROE). The results show that SIZE and LIQ impact negatively on CAR with significant meanings. On the other hand, LLR and LOA also affect CAR negatively but they are insignificant.

Estimating IBNR claim counts using different levels of data aggregation

Reserving problem in non-life insurance is an applied statistical problem with several computational aspects. In the literature, the focus has mainly been on aggregate reserving techniques and the chain-ladder method with its extensions has remained as the most widely applied claim reserving method. The classical chain-ladder method is regularly applied to annual data, but the question arises whether the reserve estimates based on more refined data outperform results obtained by annual data. We investigate whether and how much different data aggregation levels can improve the reserving process. To compare the performance of the classical chain-ladder method and its novel continuous extension, we conduct two simulation studies as well as a case study with an insurance data and use both models to estimate the IBNR claim count estimates on different levels of data aggregation. The results demonstrate that the continuous approach of the chain-ladder method provides in general a minor improvement in the IBNR predictions but proves to hold its predictive power on different types of data. This study highlights data aggregation levels in which the estimates of classical chain-ladder model outperforms the estimates obtained by the use of annual data, providing more insights for establishing accurate loss reserves.

Современные подходы к оценке и выбору вариантов погашения проблемной ссудной задолженности

The article suggests the significant proportion of bad debt to be one of the main problems of the modern bank lending system in Russia. In banking practice, the debt of borrowers is considered problematic if there is a delay in the principal debt and / or interest during 60 calendar days or more (the last 180 calendar days). The presence of bad debt requires creating reserves for possible losses on loans, which can reduce the financial result and the loan portfolio quality 
 of banks, so paying off bad loans is an urgent task for commercial banks. There is no single algo-rithm for choosing the best way for debt repayment. There has been analyzed the research of evaluating the value of claims and calculating the maximum amount of debt repayment. The factors determining the existing structure of problem debt repayment in banks have been investigated; factors 
 of assessing the effectiveness of debt repayment by concluding assignment contracts have been considered. The author's approach to calculating the financial result and cash flow under different ways 
 of fulfilling the obligations of borrowers has been proposed. The calculations and comparison of variants to pay off a debt of a specific borrower on the basis of a financial model have been made. Due to the divergence of cash flows, the assignment of claims to the bank is found the preferred method for debt repayment. It has been stated that the assignment of claims has several advantages over other methods of debt repayment: it helps to save the lender from the costs related to the repayment of bad loans; from the need to reserve additional funds provided for compensation of possible losses; replenishment of working capital, partial coverage of their losses, improvement of official statistical indicators by reducing the share of problem loans; concentration on their business.

Tournament Incentives and Reserve Management

This paper examines the relation between tournament incentives and reserve management. We find a positive relation between internal tournament incentives and reserve errors, implying that a larger pay gap as a tournament prize induces vice presidents (VPs) to overestimate loss reserves. In other words, a higher tournament prize is associated with conservative loss reserve management. Unlike the literature, we do not find a positive relation between tournament incentive and profits (risk taking behavior). Taken together, the evidence indicates that VPs focus on strong financial health of the firm instead of its profitability. In addition, we find the impact of internal tournament incentives on the reserve error is more pronounced for larger, financially weak and more geographically focused firms, and is mitigated for the firms with higher percentage of claim loss reserve over total liability and paying relatively higher tax rates. Our results also suggest that SOX mitigates the conservative reserve behavior. Finally, we also find that as board independence enhances, VPs induced by promotion-based tournaments become more likely to have conservative reserve behavior.

Value Relevance of Add-back of Loan Loss Reserves

Value Relevance of Add-back of Loan Loss Reserves

Accounting for Financial Instruments under IFRS 9 – First-Time Application Effects on European Banks’ Balance Sheets

IFRS 9 was introduced by the IASB in 2014 and became mandatory for fiscal years starting in 2018. It bears fundamental changes in the accounting requirements for financial instruments, especially in the areas of recognition, categorisation and measurement, impairment and loan loss provision. As bank balance sheets consists of financial instruments by more than 95 per cent, the application of the new standard was expected to bear major impacts on the bank´s balance sheets. This working paper covers a review of pre-application impact assessments publishes by regulating authorities, such as EBA, auditors and researchers. Main part of the working paper is a comprehensive overall balance sheet impact with a closer look at the effects on CET1 ratios. Therefore, detailed movements between previous and current classification categories as well as the portfolio composition per category are analysed. A specific look on the classification of equity instruments as well as potential changes in maturity of long term investments in loans and bonds are taken. It can be shown that the equity investments usually are not specifically categorised and that IFRS 9 doesn’t have an effect on long term investments by banks. The study also outlines the transition impact from the incurred loss to the expected loss model with a focus on the allocation of assets to the newly introduced three impairment stages and the development of loan loss reserves. The data of the study is clustered by characteristics to show the influence of size, credit risk approach, business model and country and tests the findings for statistical significance. That allows to outline visible trends affecting the new standard´s impact severity.

DeepTriangle: A Deep Learning Approach to Loss Reserving

We propose a novel approach for loss reserving based on deep neural networks. The approach allows for joint modeling of paid losses and claims outstanding, and incorporation of heterogeneous inputs. We validate the models on loss reserving data across lines of business, and show that they improve on the predictive accuracy of existing stochastic methods. The models require minimal feature engineering and expert input, and can be automated to produce forecasts more frequently than manual workflows.

Property-Liability Insurers’ Discretionary and Nondiscretionary Loss Reserve Error: Relation with Investor Sentiment

We examine whether investor sentiment is associated with loss reserve estimates of property-liability (P/L) insurers. Using the Michigan Consumer Confidence Index as a proxy for sentiment, we find that the level of investor sentiment is negatively associated with discretionary component of loss reserve error. In contrast, our evidence does not suggest a similar relationship hold for investor sentiment and nondiscretionary loss reserve error. Further analysis indicates that stock insurers are more sensitive to investor sentiment than mutual insurers, in terms of discretionary component of loss reserves. The results are consistent with our hypothesis that P/L insurers cater to investors’ optimism (pessimism), driven by investor sentiment, via discretionary loss reserve claims. For robust test, we also measure investor sentiment by using two alternative proxies: the Conference Board Consumer Confidence Index, and the index in the stock market developed by Baker and Wurgler (2006, 2007). The results are consistent. Our study discovers a new rationale for why insurers may use discretion over their loss reserves.

Penalising Unexplainability in Neural Networks for Predicting Payments per Claim Incurred

In actuarial modelling of risk pricing and loss reserving in general insurance, also known as P&C or non-life insurance, there is business value in the predictive power and automation through machine learning. However, interpretability can be critical, especially in explaining to key stakeholders and regulators. We present a granular machine learning model framework to jointly predict loss development and segment risk pricing. Generalising the Payments per Claim Incurred (PPCI) loss reserving method with risk variables and residual neural networks, this combines interpretable linear and sophisticated neural network components so that the ‘unexplainable’ component can be identified and regularised with a separate penalty. The model is tested for a real-life insurance dataset, and generally outperformed PPCI on predicting ultimate loss for sufficient sample size.

Analisis Perbandingan Efisiensi, Kualitas Aset, Dan Stabilitas Bank Umum Konvensional Dan Bank Umum Syariah Yang Melakukan Pemisahan (Spin-Off) Di Indonesia Periode Tahun 2013-2017

This research aims to compare the efficiency, asset quality, and stability of Islamic banks that do spin-off and conventional banks in Indonesia from 2013 until 2017. The variables used are overhead cost and cost income ratio as efficiency proxy, the loan loss reserves is a proxy of asset quality, and return on asset and equity to asset ratio are proxy of stability. The sampling technique is total sampling. The population in this study amounted to 22 banks. The statistics test used are descriptive statistic, Mann Whitney test, and Independent Sample t-Test. The results showed that there were no significant differences in efficiency and asset quality, but stability was significant difference from 2013-2017.

Determinants of capital adequacy ratio on banking industry: Evidence in Indonesia Stock Exchange

Capital adequacy ratio (CAR) is an important indicator of bank safety sustainability. Banks that can guarantee CAR means the bank has the power to resist the financial crisis, protecting the bank itself and funds from depositors. This study aimed to determine the factors that affect the CAR. The sample used in this study is the banking industry listed on the Indonesia Stock Exchange (IDX) from 2007 until 2018. Independent variables are bank size, leverage, loan loss reserves, net interest margin, loan assets ratio, and liquidity. The dependent variable is CAR. The number of samples is 27 conventional banks by using purposive sampling. By using panel data regression analysis by estimating ordinary General Least Squares (GLS) method. The results of this study indicate that bank size, leverage, loan loss reserve, net interest margin, and loan asset ratio has an effect on CAR significantly while liquidity has no effect on CAR. The results of this study are expected to be used as a reference for bank managers and investors in looking at the factors that affect the CAR in the banking industry. JEL Classification : C33, G21, G30 DOI: https://doi.org/10.26905/jkdp.v23i3.2981