AY1516 T2 Team Hew - Overview/Interim Review

Home

Team

Overview

Project Management

Documentation

Proposal

Interim Review

Final

Revised Project Background

Initially, it was mentioned in the proposal that Tokio Marine Life Insurance Singapore was to provide us with their data for analysis. However due to unforeseen circumstances, they were unable to extract and anonymize the data in time. After discussing with our Project Sponsor (Benito Mable), we will be focusing on another dataset which was supplied by TMI as of end January 2016. Objectives were subsequently revised with our Project Sponsor to reflect the different nature of this new dataset.

Revised Motivation

Tokio Marine’s Group Companies (GCs) collect a lot of data required for underwriting products only at the time of sale. Over time, many data points have been captured with little insights derived other than for underwriting purposes. This data is stored on multiple platforms. While some customers have multiple products, it currently is limited in the utilization of data captured to really understand the profile of the customers, what they bought, channel preference, etc.

Tokio Marine’s Asian GCs have been participating in a large-scale regional project, in which the various GCs are undergoing a phase of digital transformation and to stay updated with current technologies. Tokio Marine Asia seeks to convince staff and implement the usage of analytics among the GCs. This project serves as one of the pilot initiatives, with one Asian GC participating - Tokio Marine Insurance Indonesia (TMI). This project aims to use the insights gathered to formulate new marketing initiatives or product ideas.

Data

There will be 2 original datasets provided by Tokio Marine Insurance Indonesia (TMI). The first dataset (“motor_policy30”) contains about 2 million motor insurance policy transaction records, and the second (“motor_claim7_combined”) consists of about 600,000 motor insurance claims transaction records. Both datasets span from 2003 to 2015, and are specific to customers residing in Indonesia only.

The first dataset (“motor_policy30”) has about 152 variables, and the second (“motor_claim7_combined”) has 66, all of which will be included in Appendix A of the Interim Report for reference. Full disclosure of the dataset will only be available to parties which have signed a Nondisclosure Agreement.

The data is of a transactional nature, where it follows a hierarchy such that in the “motor_policy30” dataset, each Policy can have one or more Risk_NO. These in turn correspond to the “motor_claim7_combined” dataset, where each Risk_NO has none or many Claim_NO, where for each there are at least one Transaction_NO. Policies which did not have any claim made under its tenure will not be recorded in the “motor_claim7_combined” dataset.

Example Data:

Each Risk_NO represents a vehicle insured under the same Policy number, which can be referenced to a particular customer. However, the dataset does not include any customer details.

Claims Process

The multiple Transaction_NO values for each Claim_NO serves to record the transactions corresponding to various steps in the claims-making process.

The columns of ClaimOS, ClaimPaid and ClaimInc emulate the relationship of that of a balance sheet, and can be represented by the equation below:

ClaimPaid + ClaimOS = ClaimInc

Each row of these columns show the transactions applied, but does NOT give the current value of that column. To derive the actual amount of the total ClaimPaid, we can simply sum up the transactions, or use the value in the very last transaction (shown by the cell highlighted in green). This is best illustrated by the example below:

Transaction_NO = 1 corresponds to Step 3 in the Claims Process chart, where the ClaimOS is an estimate of the ClaimPaid recorded when Tokio Marine staff have assessed the damage. ClaimInc is subsequently 500,000 as calculated by the equation given. Transaction_NO = 2 corresponds to an adjustment in the estimated figure, to more accurately reflect the actual damage. Transaction_NO = 3 and 4 are the transactions recording the customer receiving the claim amount from Tokio Marine.

Revised Limitations

Limitation	Solution
Lack of customer demographic data, as the data is collected only when a claim is made	Limit certain analysis to only claims subset data
Lack of other costs like overhead costs which increases cost of each policy	Compare only gross loss ratio and not combined ratio with competitors
Complicated business relationships which might skew analysis or have counter-intuitive results	Frequent consultations with project sponsor to clarify any surprising and counter-intuitive results

Revised Objectives

1. Motor Insurance overall profitability and profitability by brand
2. Motor Insurance and claims trends
3. Analyse characteristics of Top Agents by Loss Ratio and profitability
4. Marketing recommendations to improve business performance based on our findings
5. Time series forecasting of profitability (which will only be discussed more fully in the final report)
6. Other predictive models (e.g. Multiple Linear Regression), as a bonus

Scope of Work
Due to the sheer number of records in the dataset, we will be restricting our scope to only focus on policies which were underwritten starting 2012 to 2015. Product lines like Shariah/Takaful Islamic motor insurance were discontinued in 2013 so their data will be excluded from the analysis

Research & Methodology

Literature Review

Indonesia’s motor insurance market has been undergoing dramatic shifts both in demand and supply. GDP is estimated to grow at a steady pace (6%), ushering in larger and more affluent middle class. Both Passenger and Commercial Vehicles have ~10% CAGR. Motorcycles have a positive outlook of 5% CAGR similarly. Prima Facie, this presents a huge market for motor insurance .

However, the government in Indonesia has been sending mixed signals, mostly negative, to the motor insurance industry. Unlike many countries, Third-party insurance is not mandatory and usually clients are ‘forced’ to purchase coverage because it is a condition of the leasing agreement and a majority of Indonesians require financing. These clients seldom renew their policies as it is not mandated. Ernst and Young succinctly summarizes the market condition below:

“This has caused motor insurers to focus their distribution efforts on building relationships with dealerships and lenders, such as banks and finance companies. It explains why a local insurer with conglomerate links to major dealerships, such as Toyota and Isuzu, holds a dominant place in Indonesia’s motor insurance industry.”

In 2014, the Finance Ministry of Indonesia (OJK) introduced tariffication which meant that motor insurance companies have to follow a common set of guidelines for pricing of premiums and discounts. According to consulting company Willis International, this meant that motor insurance policies are going to be more standardized and thus competition will be centred largely on pricing .

Further, in 2012, the government increased downpayment required when leasing motorcycles and cars to 25% and 30% (from 10%) respectively to prevent a credit risk and property bubble developing. This affected automobile dealers and manufacturers like PT Astra (which distributes Toyota and Honda) and Suzuki causing their share price to drop by 5% . In a dramatic reversal of policy to loosen monetary policy and boost domestic demand, in 2015 the government lowered the downpayment for passenger vehicles and motorcycles by 5% . On the positive side, the government has slowly moved to a Risk-Based Capital Framework favouring companies with large capital reserves like Tokio Marine. As such in the Indonesian market, consolidation is taking place with smaller domestic players exiting. The top 5 players have about 50% market share in 2010 .

The key issues facing Indonesia’s non-life insurance industry are navigating the regulatory environment, developing new channels of distribution (bancassurance) and managing profitability in this high volume low margin business. The key indicators to pay attention to are Loss Ratio, Average Premiums and Claims Rate and Expenditure. Although benchmarks differ according to geography and degree of competition, it helps to put Tokio Marine Indonesia’s performance in perspective. Generally, the more developed markets have lower profit margins and Multi-National companies tend to dominate. It is against this contextual backdrop that we will proceed with exploration of data.

Methodology

Exploratory Analysis

Our group used mainly JMP Pro and Tableau for the initial stages of Exploratory Data Analysis. The Exploratory Analysis was split into motorcycle, vehicle, personal and corporate depending on the variables analysed. Our approach was to look at the general patterns of our data before focusing on the details. JMP is used to do statistical analysis, plotting of treemaps as well as investigation of statistical distribution of the data. Tableau is used for analysis involving a time series as it is more effective in grouping the data by time periods and has more interactivity.

Time Series Forecasting & Predictive Modelling

Our group’s predictive model will be using profitability measures like GWP and ClaimPaid, as the dependent variable. We have omitted measures like Commision and Discount as they can be controlled by the company. Our group also aims to create predictive models for different segments like Agents, Brands and Vehicle segment. This allows the company to project which business segment will tend to be profitable.

One technique that our group will be using is the multiple linear regression method. Some of the independent variable were are looking to use in our regression model are Geo_Location, Jap_ind, Corppers, Counter1, Business_ind, Channel, Cover_Type, Area. We will be using measures like t-ratio and VIF to investigate the validity of these independent variables. Time Series Forecasting & Predictive modelling will only be explored after exploratory analysis is fully completed and the full methodology will be elaborated more in the final report

Software Chosen

The two main software that we used were JMP and Tableau. The reason for using 2 softwares was that each has different strengths. JMP was used primarily for cleaning the data and exploratory data analysis while Tableau was used mainly for visualization and creating interactive dashboards.

JMP has several features which makes cleaning data much easier than Tableau. JMP has a large variety of in-built functions to manipulate data tables like ‘update’, ‘join’, ‘split’ et cetera. Unlike Tableau, only the ‘join’ function is in-built and other more complicated conditional joins have to be written in code. JMP also has the functionality to selectively brush data points on graphs created and it is linked to the data tables as highlighted rows automatically. This is immensely helpful in identifying outliers and subsets within data. Of course Tableau can do this too, however it is more tedious as the user has to manually filter the data on the data table. Furthermore, JMP provides more in-built statistical functions like analyzing simple statistics and other statistical tests like ANOVA.

Tableau’s strength is that it has more aesthetically pleasing color palettes and more functions for creating storyboards and dashboards which JMP is unable to offer. In creating visualizations, Tableau has the ability to place local and global filters (in dashboards) in the user interface directly. However in JMP, the user has to go back to the data table to manually filter the data. This allows the end user of dashboards to interactively slice and dice the data on the various visualizations intuitively.

Both JMP and Tableau have the ability to load and export excel files thus working with these 2 programmes concurrently is not a problem.

References