Good morning. My name is Andrew Dempster, and I'm the General Manager of Investor Relations at Judo Bank. It's my absolute pleasure to welcome you to our virtual update this morning on our technology strategy and CTI. We are in Melbourne today, which is the traditional lands of the Wurundjeri people, and I would like to pay my respects to elders past, present, and emerging. This morning, you will hear from our CEO and Co-Founder, Joseph Healy, as well as our COO, Lisa Frazier, and our CFO, Andrew Leslie. The formal presentations will last for about 40 minutes, and we've left lots of time for Q&A at the end of the session. Our material was lodged with the ASX this morning and is also available on the Judo website. A replay of this session and our Q&A will also be available on the website later today.
Lastly, the presentation material and the presentation itself may contain forward-looking statements or opinions. In that regard, I will just draw your attention to the last slide of our presentation. I will now hand you over to our CEO and Co-Founder, Joseph Healy. Thank you.
Thank you. Good morning to everyone, joining us on the call this morning. It's great to have you with us and it's a great way to start the week. We're very pleased to be presenting our deep dive on technology as our technology strategy and CTI. In addition to the deep dive, we've also included a brief trading update, which demonstrates that we're on track to achieve our FY 2023 guidance. We remain very committed to providing the market with detailed information, informative disclosure to help you better understand our business and its future prospects. Later in the year, we're going to do a deep dive on our credit portfolio and our unique culture. The focus today, as Andrew mentioned, is very much on our CTI.
I'm joined on today's call by our Chief Operating Officer, Lisa Frazier, and our CFO, Andrew Leslie. Andrew is well known to most of you and perhaps Lisa less so. Lisa joined Judo in August 2021, her remit covers technology, operations, and products. Lisa also takes the lead in our strategic planning, in particular, how we bring to life our ambition of building a world-class SME bank, which has a great customer experience at the heart of our strategy. I'll shortly hand over to Lisa to take you through the key elements of what we mean by an experience-based strategy and the role that technology plays in bringing it to life.
After Lisa's presentation, Andrew will walk through our the financial implications of the investments and the flight path to our CTI at scale, as well as providing a high-level update on how we are performing in the current financial year. We're leaving plenty of time at the end for Q&A. I'm very much looking forward to hearing your questions. Before passing to Lisa, I want to recap on some of the important themes in the next couple of slides, themes that are central to understanding the Judo thesis. We are a purpose-led bank. Our purpose is to be the most trusted SME business bank in Australia. Our business was born out of a belief that there had been a market failure in the provision of SME banking services by the major banks, in particular, as they have continued to industrialize their operating processes.
That failure was evident in the numerous poor customer satisfaction surveys that have been a feature of the industry for such a long time. In building Judo Bank, we wanted to bring back the craft of SME banking, where relationships are, as Lisa will highlight, where customer experience trumps products. Being a specialist bank means that all we think about and all we talk about is SME banking. This creates a great advantage for us in both in our agility and our speed in moving and making decisions, in contrast to the sometimes slower-moving and heavily bureaucratic nature of the more complex banking system. We have built Judo into what I believe is a truly unique bank. There is no Judo Bank lookalike anywhere else in the world. I believe the uniqueness of the Judo Bank business model is evident in our financial performance to date.
Globally, no other startup has got to profitability in the timescale that Judo Bank has. In fact, there are many startups that have a longer history than Judo that are still struggling to reach profitability. A big factor in that, we'll talk about this today in the nature of our experience-based model, but another big factor has been the caliber of the management team, that is running the bank. I have never worked with a more cohesive, a more energized, and a more talented management team than we have running the bank today, something I'm enormously proud of. The achievements of the bank have also been recognized in numerous awards, I call out just a couple of these.
Recently, we were recognized in the Financial Times Top 500 Growth Companies in Asia Pacific as number four, which was an amazing achievement, which also saw us as not only number four amongst the Top 500 Growth Companies in Asia Pacific but as the fastest growth company in the financial services sector. Last week, we were also re-recognized LinkedIn as the number one company in Australia and in the AFR Best Places to Work. These are just some of the accolades and numerous accolades that we've achieved over the course of our very short history. In fact, it was only last week that we celebrated the 4th anniversary of becoming an ADI. We've done a lot in a very, very short timeframe.
I want to move on and just kind of recap on an important picture of how we have conceptualized and visualized the evolution of Judo Bank. This is a slide that you saw in the half year results back in February, which talks about the three horizons. Horizon one was all about building a bank and getting it to market. With technology that was fit for purpose, but not necessarily fit for future. This approach contrasts with other models that emphasize technology, models, both here and elsewhere in the world, that focus on technology and building technology, but not being clear on the nature of the economics and viability of the business strategy. We felt it was important that the economics we had demonstrated the economics of our model first, and build the future banking business model second.
This as I, as I say, is in contrast to the fintech fad that dominated the thinking of so many others as they launch new challenging models around the world. Horizon two is largely about investing in technology and building the scale to enable our bankers to deliver a greater experience and our customers to enjoy a greater experience from us. We are now very firmly in horizon two, executing a clear plan to build the scale benefits that are inherent in our model and were always inherent in the vision that we had for the company. Horizon three is about building a world-class SME bank. A bank that is growing consistently above system, achieving on a sustainable basis the economics that we have flagged in our market as at scale metrics that we announced to the market back in November 2021.
Also in addition to those financial metrics, delivering best-in-market customer NPS, and by some margin, as well as best-in-market employee engagement, and again, by some margin. When we look at our NPS score today, which is +66, this is in an industry that often struggles to get into positive measures. As I say, by some margin, and that's our focus. Moving on to the main focus of today's presentation, which is our CTI. At the time of the IPO, we outlined our metrics to scale, as I mentioned, which we demonstrated our confidence, which we have repeated in all of our updates since then, our confidence in achieving those goals.
To recap, the picture we painted is that we will have a business with a low to mid-teens ROE with a lending book at scale of between AUD 15 billion and AUD 20 billion, a net interest margin of at least 3%, a cost of risk within 50 basis points, and our CTI approaching 30%. The CTI target of approaching 30% is not one that the industry and observers of the banking industry are necessarily familiar with because there is no precedent for that outcome. Notwithstanding that fact, we do not see it as an heroic ambition or an heroic goal. There's two reasons for that. First of all, as you can see from this slide, we are a 3% NIM bank, whereas the rest of the sector tends to operate on a sub 2% NIM.
We benchmarked the CTIs for the rest of the sector, assuming that those banks were achieving a 3% NIM instead of 2%. As you can see from the slide, using that measure and against most recent disclosures from the other industry participants, that all of them would achieve a CTI comfortably in the mid 30% if they were achieving a 3% NIM. That's not the main driver behind our 3% NIM. The other key driver behind our 30% CTI is that we do not carry any significant legacy or infrastructure costs, the costs that our competitors are burdened with. We have designed Judo from a blank piece of paper with no legacy branch network, no legacy infrastructure, no legacy people, and no legacy culture.
We've built a modern technology platform which is cloud-based and provides us with significant flexibility as we grow the bank. Of course, the CTI percentage to the ROE goals that we've set ourselves. Another important measure for us in thinking about CTI is the quality of our costs. The incumbent banks, some of which have existed for well over 100 years, carry a significant amount of deadweight costs in their cost base. Our goal is not only to be efficient as we're building Judo to scale, but to ensure that the quality of our costs remains high, with a large proportion of our costs remaining customer-touching.
At scale, somewhere between 65% and 70% of our FTE will be in roles that touch the customer, which is far superior to other banks which pay a large tax for internal bureaucracy and complexity. Andrew will come back on to discuss this important theme in his presentation. On that note, I'll now hand over to Lisa, who will explain the technology strategy and the investments we're making, which are key to delivering our at scale CTI metrics. Lisa.
Thank you, Joseph. Good morning, everybody. Today, as Joseph outlined, I will be sharing with you our technology strategy in detail and the path to achieve our at scale metric of approaching 30% for CTI. CTI is a metric amongst many that contribute to defining Judo's aspiration to be a world-class SME bank. It is not a metric to be pursued in the absence of strategy. I will start with a refresher on our strategy, as Joseph mentioned. I will then cover our key investment areas in tech, data, and digital, explain how we will measure scale over time. Lastly, our implications for CapEx over the coming years. You have heard us speak about our strategy as an experience company. In a digital economy, there are essentially three business models: product, experience, or platform.
Successful businesses need to be good at all three, but must excel at one to outperform the competition. Banks today are predominantly product companies. Think about the advertising you see from the industry, largely communicating products and pricing. Judo is different. Our business model is predicated on the experience we deliver to customers via our relationship bankers, an experience that includes a dedicated banker that knows our SME customers and their businesses personally, an experience that customers are willing to pay for. Reducing costs and improving CTI is not new to the industry. In fact, for decades, innovation in banking has focused on removing transactions using technology. Starting with the ATM some 60 years ago to mobile app banking today, customers can self-serve 24/7 on more tasks than ever before. Convenience is great. However, the unintended consequence has been a disintermediation of banking relationships.
How close are you, the people you know, to your or their bank? For SME customers, this has led to low levels of financing, long wait times for answers to their needs, and frustration as no one seems to know them, care about them or their businesses, even if they've been customers for years and years. At Judo, achieving CTI will involve technology, data, and digital to do the following. Firstly, enable our bankers to achieve service levels that delight customers at scale. Secondly, drive efficiencies across all Judo, e.g., in operations, risk management, and finance, reducing the future FTE needs we require as we continue to grow. Let me now walk you through all three and their contributions in our path forward. Technology. Judo's approach is to leverage technology to support human relationships, AKA relationship banking, at high efficiency.
While there is a valued role for technology to enable customer self-service, it is insufficient to serve SMEs well. We are focused on using technology to enable banker and customer relationships, i.e., achieving our competitive advantage. We have a purpose-built, modern, cloud-based tech stack and leverage adoptive technology principles. As you are familiar, cloud-based modern technology offers a cost advantage, and this, combined with the absence of decades of legacy systems, means we have a lower cost to run our technology. The services that do not differentiate us, we buy and adopt, not adapt or customize. This is critical to managing technology total cost of ownership or TCO. We keep track of any necessary customizations to ensure we keep our systems clean and that they are used for the purpose they were designed for. We have another important advantage that is often overlooked.
We have a flexible tech stack. By that I mean, we can cherry-pick solutions and technology providers to best meet our business needs and continue to optimize TCO going forward. A flexible tech stack also means we can cost-effectively build what differentiates us. This is the structural cost advantage of Judo. We are not precious about our technology decisions. If there is a better mousetrap, we'll find it and use it, and it costs us a fraction to change. I will talk about this later, the move of deposit origination and our new digital app build were completed in roughly 12 months, short timeframe by industry standards. The flexible tech stack capability drives ROI on our CapEx investments and importantly, contributes to all our at scale metrics, not just CTI. Joseph described earlier our three horizons. We have updated our technology strategy for Horizon 2, scaling.
Judo Bank started with small investments in technology. Our preference was to focus our capital on establishing our relationship bank, hiring bankers, and proving our customer value proposition in market. Technology in the early days of Judo Bank was completely outsourced to numerous providers. This model was proven then, it is insufficient for scaling going forward. In fact, TCO is best optimized by a balance between the economics of outsourcing, the criticality of having a knowledge base in-house to run the bank, and the ability to build what differentiates us. To that end, we will be doing four things in technology. One, building new capabilities internally, including investing in engineering. Two, optimizing our cost of delivery, removing high-cost providers via insourcing or striking different, new, more cost-effective partnerships. Three, consolidating what is working, e.g., scaling our automated testing capability.
Four, building the banking experience that enables our bankers to substantially outperform the industry. Moving to the role of data and insights. Data is essential to our experience strategy in achieving CTI. My experience in using data is that it is an asset that is generally underutilized. Why? Well, because the data and insights are often not in the hands of the front line or decision-makers. The design for data use cases, advanced analytics, and even AI often fails as the insights are difficult to access or action. Our data strategy is founded on empowering bankers and customers. Using a cloud-based data platform built to scale to deliver reporting, insights, advanced analytics, and leverage artificial intelligence in the future, we will deliver capabilities to bankers to enable them to scale their portfolios. Here is how we will approach this for bankers.
As a banker, give me the information I need to serve customers well, e.g., application status, upcoming settlements, or even open servicing tickets. Two, as a banker, make things easier for me. Upload open data for information gathering, use automated spreading, and automate document creation. Lastly, help me prioritize where I need to focus each day, from risk indicators, price insights, customer events, or reminders. This is certainly exciting times ahead for bankers and customers going forward at Judo. Bankers are not our only focus. In fact, we must leverage data across all of Judo. We believe in a distributed data model with security that enables our teams within Judo to self-serve data and insights.
The data insights team within my group is small, and they are responsible for data manufacturing and data tools, data quality, data governance, and the advanced analytic capabilities, but they do not provide the everyday reporting. Remember my earlier point about data in the hands of the front line. The data platform and insights tools we are building enable self-service. One example is our relationship bank, who today leverage the reporting tools we have created and have built their own dashboards. This is the most effective and efficient way to supply meaningful data and insights to the relationship banking team. Thirdly, digital. Digital is table stakes. It is important to customers. I am pleased to share that last week we began the rollout of Judo's new digital banking experience as pictured here. Our lending customers will have digital access to their account information and product information.
Rolling out digital will remove hundreds of calls to our bankers from customers who are seeking latest balance information, statements, or other transactional information requests. In the coming weeks, we'll launch the new digital banking experience for deposit customers. This will enable us to retire the existing deposit digital banking experience, which will result in lower calls to our customer support team due to better reliability and better design. The Judo expense base is largely consists of people. We do not have branches in hundreds of locations, nor do we own data centers. Essentially, we are people with laptops and mobile phones. Therefore, digitization broadly is important in delivering improvements in our processes and services to enable our bankers to spend more time with customers and our support staff to focus on higher value activities. This is the focus of our future investments. Where to next?
It is important to remember our bank is young. Yes, we are scaling, but we are still building. In fact, there are major capabilities yet to be built. We operate with banking basic capabilities to support our processes and our customers. We will be introducing proven technologies such as CRM, workflow, and ERP to reduce future cost growth. These platforms have known industry benefits as illustrated here, with workflow platforms achieving up to 30% on various processes, CRM systems contributing to productivity in the similar range, and even ERP systems can generate up to 25% improvement in efficiency. To date, we have achieved our financial results with systems required to operate a bank. Going forward, we are investing in system that enables us to scale the bank. Starting in FY 2024, technology for scale will be the focus of new investments.
Collectively, these investments will help scale our banker portfolios and will slow down future recruitment needs in areas such as operations, finance, and risk, while our lending and deposit books continue to grow. Together, these investments are a major contribution to the reduction of our expense growth rates. While we have new areas to invest ahead, it is important to recall we have achieved much in recent times with the build of our existing capabilities. In fact, over the last two years, we have stepped up investments to scale Judo. These investments included core banking with lending in 2021, account origination with the introduction of nCino, the data platform, and information security. Even though these capabilities exist today at Judo, in line with our flexible tech stack strategy, we continue to evaluate providers over time and will make changes appropriate to our business.
Let me share the benefits of our two most recent investments. The first example here is leveraging nCino for origination of our deposits. We introduced nCino for our lending applications in 2021. In short order, we were able to see the benefits of the platform in lending. We made the decision to follow fast with deposit origination. Since go live in November 2022, we have originated AUD 1 billion in deposits. We've added new products, SME FTDs and business TDs, and we've improved complete rates. We've reduced our operational costs with less calls down by 80%. We've achieved a higher net promoter score by nine points, which by the way, is in the sixties, and we lowered the technology team time used to run origination. Awesome outcomes. The second area example is around the new data platform.
Back in early 2021, when I started at Judo, the team and I completed an assessment of the scalability of our tech stack. The one platform we determined that would not scale was our data platform. The architecture build and run costs would not meet Judo's future needs. We have built a new platform suitable to grow and scale with the business. From the early days of Judo, our data platform was in AWS and built and run by external resources. We decided to completely overhaul the data platform at a cost of just AUD 2.8 million and bring the build and run capability costs in-house. We now have a single reporting data platform running in Azure, enabling our business teams to perform analysis, gather insights, and meet regulatory requirements. Total cost of ownership of the platform has come down by 50%.
We have higher reliability and a lower cost for future build. The platform is in flight to be leveraged for operational data capabilities, as of March, we used it to meet open banking requirements. We're very proud of this outcome. We will continue to invest in upgrading these earlier four investment areas as they contribute to all our at scale metrics. I'm gonna shift gears now and talk about scaling. How will we know we are on track to scale? Good question. When scaling businesses, measuring what matters really does matter. My time with previous startups in San Francisco demonstrated time and again the importance of unit economics as the measure of scale. Without a compass, CapEx investments and OpEx costs can grow, not necessarily have the right balance of revenue versus profitability.
At Judo, over the last 12 months, we have put in place unit economics, our scale measure. Let's start with a short introduction to unit economics. Essentially, unit economics measures the marginal cost to manufacture a widget. In our case, a widget is an SME loan. We detail top-down and bottoms-up all the activities that contribute to the cost to acquire, originate, settle, and service a loan. Here is unit economics in action. At Judo, an average loan requires 60 hours to originate and 30 hours to service. The chart illustrates how unit economic measures the time spent across roles at Judo supporting an SME loan. Our bankers spend the majority of their time on business development and credit assessment, but also spend time servicing customers, all good activities.
Measuring unit economics help us optimize where bankers, analysts, credit, and support staff should ideally spend their time in the future. Over time, our unit economics analysis tells us two things. One, where to invest to have the highest impact on CTI, i.e., identify the largest areas for scaling up the bank. Two, tells us how well our previous investments are performing in reducing CTI. Here is now our most recent set of opportunities identified. As you can see, several opportunities exist for bankers in business development, building the application, collecting information, reducing rework, and streamlining operational support. In servicing, we also see ways to reduce banker time spent. Having the digital bank capabilities I mentioned earlier is one of them. Improved covenant management, streamlining annual reviews, and simplifying our variations processes. We have seen our previous investments are contributing to scaling Judo.
I mentioned earlier digitization and process optimization are key contributors to CTI. Here are two examples of how unit economics have proven we can reduce time spent and improve quality. The first example is settlements. From unit economic analysis in August 22 and again in February 23, we saw a 15% reduction in the unit economics of settlements. The example here refers to document production and data automation. We leverage multiple legal partners to prepare our loan documents. The process relied on email and checklists, a largely manual effort. With the implementation of the LIXI API standard, our loan data is now uploaded to our legal partners automatically, reducing time spent, costs, and error rates. In fact, our bankers receive back one hour and our ops team up to 45 minutes per every loan. The second example on the right is related to servicing.
From the unit economic analysis over the same period, we saw a 23% reduction in the unit economics of servicing, largely driven by the new statement automation capability we built. An example in build for this month is our annual review process, which has not kept up with the business. We have streamlined the annual review process and improved reporting and tracking. Annual reviews will be completed in shorter time. Bankers and analysts will save 30-45 minutes per review and more time staying on top of what needs to be done each month. This is just the beginning. We are also piloting a use case with open data to support annual reviews, potentially shortening the time to complete the review to evaluate a customer's business performance. Let's move to what all this means for total CapEx investments.
Taking a five-year view, starting with the IPO FY 2022, the chart on the left breaks down the expected mix of our CapEx investment by area of focus. We will continue to make ongoing investments in the foundations layer, in core banking, origination, and information security. This involves platform upgrades and developing new tools and capabilities on these existing platforms. Digital and data will grow their share of investment as well. This incorporates advanced analytics and adding new features to digital. We will continue to meet our regulatory and compliance commitments. It is of note that our investment is lower relative to other banks, as we do not have large remediation programs, nor their level of product complexity in our business. Simplicity is Judo's best friend, I like to say.
As I mentioned earlier, starting in FY 2024, we will invest in the new capabilities around scaling, which includes CRM, workflow, and ERP. Our CapEx spend will grow from FY 2022, which was a lower investment year. We are planning for FY 2024 to be a catch-up year before reverting to the $20 million range per year. We believe our $100 million CapEx investment envelope holds. We are investing for scale, which supports our CTI metric achievement. Each of the investment areas described here help to change the structure and mix of our cost base at scale with a focus on quality costs, i.e., the banker customer experience and support, while reducing the expense growth rate in corporate functions. The Harvey balls here describe the expected impact over the years to come.
In summary, new investments in productivity systems will help to minimize manual tasks, improve the efficiencies of our analysts, our operations, and corporate functions. Enhancements to our existing technology will support the growth of our business and enable us to redefine and optimize existing processes and reduce the overall total cost of ownership of technology. Investing in our data strategy impacts all teams as we give our people the information and insights they need to prioritize their time and drive their own effectiveness. Digital will focus on enabling customer self-service of low-value tasks. It will also help our bankers service customers at scale with their portfolios. Lastly, regulatory and compliance investment will also allow us to continue to meet our commitments efficiently. All of what I've discussed today is only possible if we have the necessary team.
My leadership team is in place and has the depth of experience to execute on our technology, data, and digitization strategy while continuing to build, to expand, and grow the bank and run the bank day to day. As you can see, Euan Walker, Pete Vermeulen, Ada Caguin, and Michael Collins are all seasoned executives with deep functional experience to lead the key elements of our tech, data, and digital strategy and protect us as we grow. Combined with our product, operations, and strategy leaders, Patrick Nolan, Ryan Marchant, and Kate Schade, we are the team that designs, builds, and runs Judo's engine. Over the last two years, we have recruited across tech, data, and digital. While recruiting was slower in FY 2022, the availability of talent in FY 2023 is improved.
We have a few specialist roles to place and a number of roles to support our technology strategy. With that, I will close, and thank you for your time. I look forward to your questions, and I will now hand to Andrew.
Thank you, Lisa, and good morning, everyone. Today, I'll be providing some details on the building blocks of our at scale CTI and some color on our cost base today and how we expect it to evolve. I'll then close with a short trading update on our financial performance year-to-date. I want to begin with some context to our CTI journey. At the time of our IPO, we disclosed to the market our at scale metrics, including our CTI target of approaching 30%. Since that time, we have grown significantly and our CTI has halved from 98% to 53% today. Our loan book is now halfway to our at scale metric, and we are a profitable and growing company. This demonstrates the significant impact of scale on our business.
As you can see from the chart, by far the most meaningful driver of our CTI going forward continues to be scale. A AUD 20 billion loan book is not the end of our story. Our CTI will continue to improve as we grow. The investments and initiatives that Lisa has outlined, coupled with our significant total addressable market and our market-leading customer value proposition, will provide us with the platform to continue to grow and deliver world-class economics. As this slide outlines, there are other revenue factors impacting our CTI too. Other operating income and other net interest income movements, which includes normalized funding costs, lower liquids and channel mix, are other factors that will impact the revenue side of the CTI metric. As a high-growth bank, cost movements are comparatively small drivers of our CTI trajectory.
Whilst we'll continue to invest in customer and banker costs and tech and data, support, risk, and corporate function costs will remain relatively neutral to CTI. I'll unpack these components more on the following slide. Despite the impacts of recent inflation, we believe our at scale CTI metric of approaching 30% remains an appropriate target in light of our investment plans and the benefits derived from scale. We are growing into this target, not shrinking towards it. I want to turn now to the composition of our cost base. We have a high-quality cost base that will increasingly be structured to deliver a world-class customer and banker experience, while also keeping the bank safe, well-run, and well-regulated, and helping us maintain our competitive advantage in customer relationships and service quality. Clearly, as a growing bank, we will continue to invest, with costs increasing as we scale.
While costs will grow next year, in part due to the run rate impact of the investments we've made in FTE this year, the cost airplane is leveling out. We are nearing the peak level of staff we require and won't see the same rate of growth in FY 2024 costs that we have in FY 2023, with single-digit cost growth to follow in future years. As mentioned earlier, investments in our customer and banker experience and tech and data will continue to grow to support scale. Let me just spend some time on these two components. Firstly, on customer and banker experience, the growth largely represents spend on our banker workforce, rising to about 200. At that stage, we expect the efficiencies from our investments to streamline a lot of manual processes currently performed by our analysts, enabling them to focus even more on customer relationships.
This will, we hope, see many of our analysts promoted to bankers, with nine successfully promoted during our short history. Our primary goal here is to continue to invest in our leading service proposition. Secondly, on tech and data, as Lisa highlighted earlier, there is still some growth yet to occur in the near to medium term as we finalize building these teams and continue our investment plans. As we grow, licensing and vendor costs, as well as CapEx amortization, will increase commensurate with our investments. The investments we're making will enable us to scale with modest net cost increases in other areas, with the absolute mix of support and corporate functions reducing. Overall, the return of the investments Lisa has outlined are significant as we put in place the platform for long-term growth. Turning now to the quality of our costs.
As a reference, we believe somewhere between 15% and even up to 30% of typical bank sector costs do not apply to us and are not value-adding from our perspective. This includes a number of elements. Firstly, branch and ATM occupancy and FTE run costs. Secondly, investment spend on fixed types initiatives, customer and system remediation, and reg and compliance costs applicable to larger banks. Thirdly is what we call the complexity tax inherent within the rest of the sector. This includes costs to maintain legacy systems and products, duplicative teams, and costs of operating in multiple segments and jurisdictions. We've also benchmarked ourselves against other successful challenger banks in the U.K. and see many similarities with them.
For example, the portion of spend on technology and data is very similar, these challenger banks in the U.K. have successfully operated in the high 20s-30% CTI range, which is a helpful benchmark for our approaching 30% CTI metric. I'd now like to provide you with an update on our financial year-to-date results. For the year to 31 March, we delivered a profit before tax of AUD 87 million and annualized ROE of 5.5%. This reflects a PBT of AUD 34 million in the quarter, 27% higher than our first half average. We are very pleased with this result, which has been underpinned by our strong underlying NIM in particular. At 3.57% year-to-date, that's marginally higher than the first half. This translates to a reported NIM of 3.29% year-to-date.
Against this backdrop, CTI has again improved to 53%. The margin on new TDs for the quarter was 68 basis points over BBSW. As we called out at the first half, the funding environment is competitive. We are in the TFF refi window for the sector, and volatility has increased given events offshore. Despite this, we have continued to attract above-average deposit flows, growing our deposit book by over AUD 200 million since mid-March. Lending growth in the third quarter has been good. We have a large pipeline of AUD 1.5 billion and expect a few big months ahead as we head into the June year-end period, which typically sees our largest lending volumes. Asset quality remains benign, with no negative underlying trends emerging. Overall, 90+ days past due and impaired is still at low levels and is in line with the sector.
Whilst there's been some growth, this reset represents a handful of customers across a range of sectors with no discernible trends. Finally, CET1 ended at a very strong 19%, which is now reported under the revised APS 112 standard. Now to guidance. Given the strong underlying margin over the quarter, we now expect second half 2023 underlying NIM to be within the range of 3.3%-3.5%, which is higher than the previously guided 3.1%-3.3% range. There's been a few factors driving this change. Firstly, TD margins were lower than anticipated through the first part of the quarter. Note that TD margins on new originations in April have now increased to the top end of our 80 to 90 basis point guidance range.
This range, however, is the long-run average price range that branchless banks have paid to be number one in the TD market and is the assumed margin embedded in our at scale NIM. In addition to the impact of TD margins, we've also managed the timing of some treasury activities during the period, including wholesale debt and hedging, which helped support our NIM. Of note, our underlying NIM performance during the quarter was achieved without additional TFF self-securitization, which remains at 80% or AUD 2.3 billion. We remain mindful of TFF repayment and are taking an active approach to treasury management, balancing funding, liquidity, and NIM.
As we've talked about previously, we continue to progress our warehouse optimization program. I'm pleased to confirm that we have now reached our targeted $2.5 billion of committed warehouse capacity, having recently signed agreements with two financiers for an additional $750 million of capacity, with pricing of those new facilities in line with existing facilities. On expenses, we reaffirm guidance provided at the first half. On cost of risk, we've had no write-offs this quarter. The book remains in great shape. We retain our prudent approach and are reconfirming our FY 2023 cost of risk guidance to be within the $50 million-$60 million range for the year. Other than upgrading our underlying NIM guidance, we reiterate all our other FY 2023 targets. Over to Joseph for some closing remarks.
Thank you. Thank you, Andrew. One of the things that we feel very passionate about, we feel is very important in understanding the Judo thesis, is to understand the nature of being a pure play specialist bank, a pure play SME bank. We believe that in all walks of life, specialists outperform generalists. In an organization that's solely focused on one segment of the economy, deeply experienced management team, an organization that all that it thinks about and talks about is how do we build an SME bank that is seen as the best, not the biggest, but the best in the market. As Lisa underlined in our presentation, our relationship centric model is essentially an experience-based model and not a primarily product or a platform-based model. A loan from Judo is ultimately no different from a loan from any other bank.
The experience in getting that loan and the service that follows the loan are the fundamental points of competitive advantage for Judo. This is reflected in our NPS score at +66. It's also reflected in the strong above-system growth, whilst maintaining sound risk/reward disciplines and in the growth trajectory that we've outlined in our metrics of scale. That whole experience-based thesis to the nature of our competitive advantage is something that we do not take for granted and is something that we will continue to invest in. It's at the core of our strategy. Other banks emphasize technology and products. As Lisa mentioned, we emphasize customer experience. We believe it is so important as a point of differentiation.
Our model is predicated on having amongst the best SME business bankers in the country. We go through and we maintain a rigorous recruitment process of interview, extensive interviews, plus a credit assessment test for those who are planning, thinking of joining the company. The pass rate on that credit assessment test is still below 50%. That has been for the last three to four years, and we will not compromise. The average experience of our relationship bankers dealing with our customers is 15 years. The average experience of the management team running the company is over 200 years of banking experience. This is a young company with deeply experienced people throughout. Our relationship-centric model is supported by our cloud-native technology platform, as Lisa mentioned.
We are excited about the potential for developing that platform, as Lisa outlined in her presentation. The significant operating leverage that we have in our in our offering is reflected in the legacy-free nature of our business and the commitment that we have to continuing to invest in our relationship banker network, which, as you saw on the slide that Andrew the at scale slide that Andrew just showed, we are still on track and plan to have 200 relationship bankers in the field by the time we get to scale. With technology there to really ensure that they are focusing on the right things in terms of our customer experience. The key message here is the power of specialization and its ability to outperform the generalist all the time.
This is a big focus for us, as well as a point that Lisa emphasized in keeping our business model simple. Avoid complexity and the complexity tax that Andrew referred to. It is something, again, that we think a lot about. The Judo strategy and its core underpinning, in my view, have a remarkable clarity, a remarkable simple nature, and a remarkable focus. Let me now just draw some concluding remarks before turning over to Q&A. We continue to execute our strategy successfully, a strategy that is clear and simple. It's the strategy you could explain to anyone, and that is understood even by those that have no real understanding of the banking industry. We are building and very committed to building a world-class SME bank with relationship and experience at the heart of the way that we think about the business.
Our legacy-free, pure play specialist model means we can deliver a high customer touch value proposition whilst delivering sustainably attractive industry-leading economics. We are very much in horizon two in scaling the bank, making necessary investments to support the continued growth in the company, whilst making sure that our costs, as they grow, are quality costs and not deadweight costs. Our purpose, as we think about our investments, is to make sure that the customer experience and banker experience are front and center of how we spend our money. Our at scale CTI at 30% is not something that we see as a heroic outcome, even though there is no market precedent for it. We remain very confident and continue to grow and support the SME economy and build a world-class bank, and doing so by running our own race.
This is an emphasis that we've placed on how people should think about Judo's growth trajectory. We are running our own race and in control of the way that we're growing our company. We are hugely excited about the future for Judo as a genuine, credible challenge bank. As I say, not seeking to be the biggest, but definitely seeking to be the best in the market. Thank you for your time this morning. We are really looking forward to your questions and observations. On that note, I'll hand back.
Thank you, sir. As a reminder, to ask a question, you will need to press star one one on your telephone. You will then hear an automated message advising your hand is raised. To withdraw your question, please press star one one again. We ask that you keep your questions to no more than two, but please feel free to go back into the queue, and if time permits, we'll be more than happy to take your follow-up questions at that time. Please stand by while we compile the Q&A roster. I show our first question comes from the line of Andrew Lyons from Goldman Sachs. Please go ahead. Mr. Lyons, the line is open.
Thanks, good morning. Just two questions. Firstly, a question on your NIMs. You've noted your 3rd quarter NIM has been supported by deposit pricing in the first two months of the year that then subsequently normalized. We've been hearing a lot about competition on the lending side of the balance sheet, particularly in mortgages. I'd just be keen to get a bit of an update just in relation to what you're seeing in lending spreads within the SME space. Then I've got a 2nd question, please.
Yeah. Hi, Andrew, thanks for your question. I mean, I think in terms of the lending margin, the dynamics there, we haven't really seen a material change from what we talked about at the first half. There's definitely spots of competition on the lending side. From a margin perspective, I think the dynamics that you saw in the first half really remain unchanged, which are really around the fixed rate component of our book. Also, I guess, just the, you know, the activity around the back book. Those dynamics and I think the impact of them, we haven't really seen a material change in the, you know, in the third quarter.
Thanks, Andrew. Appreciate it. Then just a second question just around your ability to compete with the majors in the usage of data. You've spoken a lot today about the investment you're making in your data capability, which does make a lot of sense. Just to what extent are you at, I guess, a structural disadvantage versus your peers in relation to accessing data, both from the perspective of your relative scale, but also just due to not having a transaction banking offering and what might otherwise be some cashflow insights that you would garner from having such an offering?
Great. Thank you. Look, I think there has been a lot of talk about the value of a transactional account and the data that underpins it, but we don't see that as value trade-off because of the access that our customers give to their data and now on top of that with open data. In fact, I actually would turn it around to say we have a cost effective benefit because we actually don't have all the data infrastructure required to support the transactional banking account. Time will tell though. This is early days for open data, and we are working with our bankers on using it. That in essence, I think actually leads to a cost advantage for us.
Also, we have such a small team working very closely with the bankers and other parts of the organization. Our agility to design data use cases and get them into prod is actually much faster. I mean, I built and ran the AI platform at Wells Fargo, you know, this context for us at Judo, we are very fast compared to big banks.
Lisa, I'm gonna add a footnote to your comments, 'cause it goes back to the whole relationship model. It's important that people remember that we run a very low ratio of customers to banker. We're currently sitting at around average of 25 customers per banker. Over time, that will grow, not into the hundreds. That means there's a high touch and deep connection with the customer. Relative to an industry that a lot of SMEs are catered for out of contact centers, but also even where they are looked after by a banker, that banker can be looking at 100+ customers. I would say that our high touch model provides a much stronger customer experience, but also a much stronger risk management focus than that more than compensates for any history of data collection that might perceive to be an advantage but held by others.
Thanks so much.
Thank you. O ur next question comes from the line of Jonathan Mott from Barrenjoey. Please go ahead.
Thank you. Just got two questions, if I could. The first one again on the NIM. I just want to make sure I'm understanding it correctly in the read through into the outer years. Really, it was just a timing difference on the TDs where you've made more money, in that sort of months of March, because pricing hadn't pushed up to that 80, 90 basis point range you previously talked about. Now you're expecting to stay in that range. It won't have a benefit that will flow into FY 2024. It's more a timing benefit into FY 2023. Then I've got a follow-up question.
Thanks, Jon. Yes, I mean, you're correct. This is very much a timing issue in terms of those TDs. I mean, we called out at the first half that we were seeing more competition, and we were seeing some repricing. We were expecting that those TD margins would settle in that 80 to 90 basis point range. It's really just been the timing of how that has actually played through in reality for the quarter. We have kind of seen a bit of a benefit of that, which obviously we've banked in the result. I think from a go-forward perspective, you know, we are now originating TDs in that range. Things do remain volatile, but our expectation is that that's where we will be.
It is that range that's, you know, baked in, if you like, to our at scale NIM target of above 3%. There's kind of no really no real change on that.
Thank you. If I can have a follow-up question, if I could, just on the unit economics that Lisa was talking about, the cost an hour to originate a loan. You gave us quite a bit of detail, but there was no breakdown between the cost to originate a service, a broker-originated loan versus a proprietary originated loan. Two parts to the question. Number one, how much cheaper is it to originate a broker loan, given hopefully the broker is doing a lot of that work in collections and helping you out there? If it's similar, wouldn't that imply, given there's lower revenue because you've got to pay the brokerage out, that brokers are higher CTI and lower ROE? Can you break out that cost of originating a broker versus proprietary loan?
The unit economic analysis to date has not actually differentiated between a broker loan and a direct loan. We've done this as batches of loans as they come through on the unit economic analysis in our time periods. We can look to do that going forward, but at this stage, we haven't done that type of analysis yet. I mean, the brokers represent an important channel to us, as does our direct. We will continue to use unit economics to optimize both going forward. We see differences because as you said, the brokers do provide services around the information they provide around their customers that is valuable to us. Our customers sometimes, you know, have got good relationships with their bankers, and they prefer to go direct. Either way, we'll continue to serve both direct and broker channel.
I might just add a footnote again, if I can, Lisa, 'cause this is something obviously that we've studied carefully. We can say that there's no difference credit-wise between a broker originated loan and an organic originated loan. We've studied this. Our risk, second line risk management team has studied this at length. The 70%-75% of all our, of our lending, including our organic, will come as a refi outside of another bank. They tend to be what the organic tend to be well prepared in terms of our information needs based on their previous experience in meeting the requirements of other banks.
I keep on saying in forums like this that we continue to decline about 40 %-odd of all applications that come into Judo at right at the very beginning. That's gone up slightly in recent months. There's no difference. We've always taken a view that at scale, that the organic channel will marginally be more attractive to us relative because you're avoiding the costs and you're getting some of the benefits of the unit economics that Lisa has talked about. At scale, we see that at the margin, a direct originated loan will be more attractive to us than a broker originated loan on the assumption that the risk profile remains this agnostic to which channel.
Yeah, you haven't worked out exactly how much that is at the moment by breaking down the unit economics of each channel to work out what the ROA of each channel is.
Not yet. That hasn't been the focus in the first, 12 months of the establishing the routine and the analysis. As you mentioned earlier, it is really detailed, and it is very thorough. We started and will expand over time.
Thank you.
Thank you. I see our next question comes from the line of Brendan Sproules from Citi. Please go ahead.
Hi, good morning, team. I have a couple of questions. Firstly, just on the name, just margin, I want to follow on from Jonathan's question, around the timing issue. I think you showed us in the first half that your average TD margin hedged over the one-month BBSW was sort of 47 basis points, and you're now telling us that front book pricing of that is sort of at the top end of that normalized range. Going forward, should we think about a sort of 40 basis point difference between the back book, which is still a bit of benefit for you in this quarter, versus the front book, which has moved to the top end of the range?
Yeah. Thanks for the question, Brendan. I mean, your observations, I think around where we've done that new origination on TDs is correct. Part of, I think, what we're kind of seeing through is obviously the, is when some of those deposits have actually kind of washed through in terms of supporting the underlying NIM or the NIM for this, you know, for this result, for this quarter. It's a bit of a story of how the front book and the back book on the, on the TDs is rolling through.
We will see from this NIM point that we've printed for the third quarter, we will see that, you know, prima facie with other factors. We will see that reduce as we, you know, as all of the book reprices per our expectation, our previous guidance in that 80%-90% range. There will be a component of that, you know, that does see the overall NIM reduce. The play through of that, you know, is obviously impacted by the mix of those deposits as the total funding stack evolves. Also, you know, how I guess the timing of when those were written actually comes through. There will be a reduction.
You know, we, you know, we're expecting that as we get from, I guess, this point to the at scale, you know, 3% plus NIM, that those deposits, or the repricing of the current book all the way through is gonna, you know, take a bit off in terms of, in terms of the NIM. You know, there's room to do that, I think, versus the print and the at scale level.
Can I add another footnote?
Sure.
He was provocative to jump in. Look, I think it's important to keep in mind the extraordinary environment that exists around the deposit market right now. Because of the problems that we saw offshore with the U.S. banks and some of the European banks, the refinancing of the TFF. We are seeing exceptional levels of competition in the deposit market, the economics of which are not sustainable for a lot of the major players. You know, the key for us is to make sure that we're managing through this period, I mean, protecting our economics, which we are doing, given where we are priced, made the assumptions around our pricing. You know, we see that the competitive pressure on deposits to be a phase that we're going through now that will not sustain itself.
It may be six months, nine months, but this is not a resetting of deposit pricing in the banking industry, because that would have a significant impact on the profitability of the sector, and I don't think that'll happen.
Okay. Thank you. I've got a second question. Just on asset quality. Just trying to understand, obviously you're a relatively newer book that's starting the season. I noticed the impaired assets of 41 basis points, I think, for the financial year-to-date, and that's up from 20, I think you said at the half. What sort of level of impaired assets as a percentage of the loan book would be sort of the normal level for this business? Now, I know you've got a metric of scale of the cost of risk of 50 basis points, but how would we sort of equate that into impaired assets? What would be the normal level once your book fully seasons that you would expect sort of through the cycle?
Yeah, no, thanks for the question. I mean, a couple of observations and comments on that, Brendan. I mean, as you've noted, yes, impaired's, we've seen an increase in the quarter. 90 days- plus is stable. Kind of overall, you add those in aggregate, and I think it's important to look at that in aggregate as an overall measure of asset quality. You know, that is at about 63 basis points, which, you know, is pretty much bang in line with the sector. There will be some movements in and around these buckets as things season.
The other thing, the other factor I think is just of note for the quarter is whilst the impaireds have increased, you know, there hasn't been any write-offs as well. All of these things, you know, need to settle to a normalized level. Kind of in saying that, you know, we haven't called out specifically where we would see an at scale, or a through the cycle level of impaireds. I mean, I think the way that we, and the guidance that we've provided, of the 50 basis point cost of risk, at scale is really the best measure only because, you know, it can be compared to, you know, to sector data. That's really the number that we focus on managing.
How things progress through the different buckets is, you know, is a factor of timing as much as anything else. I think the real kind of point and takeaway on asset quality is, whilst we've seen a bit of movement, you know, those overall levels are with, you know, within what we would expect when we benchmark against the sector. Importantly, there's been nothing that we've seen in the book that's led us to have any change in terms of the cost of risk guidance that we are reaffirming for, you know, for the full-year of AUD 50 million-AUD 60 million.
Yeah. We do know that historically, going right back to the 1990s, that the average over that whole cycle has been around 30 basis points. While we've said 50, that's just taking a conservative approach. I would be very surprised if we find ourselves operating at that level at scale, but that's what we've made the assumption.
Mm-hmm.
Again, people keep on saying we're a young bank, and we are a young bank, but our customers are mature businesses in the main. I mean, 70%-80% of the business we've written over the last 4 years has been from dissatisfied customers of other banks. We're very vigilant, particularly in this environment, to make sure that we're not taking the problem customers of other banks. Just given the industrialization of the banking operating model as it relates to SME and that poor touch that has been a feature of the model, that we find customers that have said, either through a broker or through networks, that actually think about Judo as a specialist bank that really understands SME banking because has, as an experience-based service proposition, as we've talked about.
We get those good businesses coming to us, and they have got long histories of being supported by banks over time. I think that's something that we will bring out when we do our credit deep dive later this year. If there is a myth that we've got a young book, we've actually got a book full of very established businesses with strong credit track record. We need to kind of inform the market much more of that. We don't have any industry exposures. We're not like a Silicon Valley Bank with a heavy concentration in any particular industry or a geography. Geographically well-diversified. We've always said that we want to build a credit book that looks like the economy, you know, with the right kind of weightings.
Some sectors which we won't have no appetite for, but broadly that we were building towards a portfolio construct rather than sleepwalking into whatever that construct might look like. This is a very purposeful, driven approach through credit risk management, both by industry and by geography. I'm looking forward to this, to revealing that information later this year, actually, because this, the book is actually quite surface full of very mature businesses, not young businesses.
Thank you.
Thank you. O ur next question comes from the line of Jarrod Martin from Credit Suisse. Your line is open, Mr. Martin.
Thank you. Joseph, just wanted to discuss the long-term assumption of a margin greater than 3%. Anyone that's been an observer of banks over multiple decades, margins go down over time. Whether, and a large component of your cost income reduction to 30% comes from the revenue side. We're going into an environment where clearly the focus has previously been on mortgages, but now we've got banks focusing on SME. Perhaps if you could make a couple of comments on why you think that 3% margin is sustainable.
Sure. The thing to keep in mind as we think about how we're gonna grow our business, even when we get to AUD 20 billion loan book, which will be circa 3% of the market. This is not a. We're not presenting Judo as a major market player. We're gonna grow by, as I've said, running our own race and being quite targeted in the type of risk that we want, consistent with the portfolio composition that we are working toward. Working with businesses that value the Judo proposition. The experience that we talked about. As I've mentioned in my remarks, I mean, a loan from Judo once banked, is no different from a loan from any other bank. A dollar is a dollar.
The experience on how you get that loan and the experience that you have once the relationship is established, is where our point of difference is. We believe that, and I know this from deep, deep experience, that customers, generally are happy to pay up to, our assumption is up to 50 basis points for that experience. That you start to lose customers when the experience fades and you industrialize. There are so many good businesses that want certainty of access to credit because they're looking to buy something or make investment decisions, and they don't wanna wait for eight weeks for a maybe yes, maybe no.
I think so long as we stay true to that customer value proposition, and given that our growth projection in the context of the market is modest, I feel that we can select the credit and customers that fit our profile and not find ourselves competing with the Main Street, if you will, the major banks on highly commoditized style lending. you know, and that's always been our ethos and through all my years of experience around SME banking, businesses generally, and I mean the vast majority, are happy to pay a premium for a great relationship. It goes to the power of specialization versus generalization. I'm very confident that we can continue to grow that and maintain our customers, even though at the margin there might be cheaper options.
With cheaper options go cheaper and poorer customer experiences, and a lot of businesses simply don't want that. At the proposition, I think has been proven. The economics that Andrew talked to demonstrate that, and, you know, we've been very consistent. We're not looking to be a big scale player. We're looking to get to scale at 20, and then there's a story beyond 20, but it's not being the biggest business bank in the country by compromising on risk/reward economics to achieve that. That, that's been a consistent theme of the way that we've talked to this business in the marketplace, that we can achieve the growth that we have set out at the economics that we believe are market leading without compromising on risk to get volume.
At that's a formula that's proven. The more and more that we prosecute into that strategy, the more and more confident I am. I don't believe that others in the industry can replicate what we do, by the way. You know, even though a lot of banks are clearly back focused on SME banking, because to your point, mortgage is no, not really seen as a growth segment. Delivering the kind of proposition that Judo as a pure play specialist bank delivers is a very hard thing for a big scale, bureaucratically encumbered organization to do in a sustainable way. There's no magic formula here. It's just doing well, the things that we believe are really important to so many SME customers, and that's our secret sauce.
Okay, thanks. Thanks, Joseph. A follow-up question maybe for Andrew or yourself. Just on competition for relationship bankers. I know you're talking much more about promoting analysts. What's the competition like out in the market in terms of being able to actually recruit relationship bankers?
The competition has been quite intense for the last 18 months or so, as certainly two of the major banks, and actually three of the major banks, have been beefing up their SME teams. That said, we are having no difficulty getting the bankers that we need. We've set ourselves a target of 200 at scale. We've got 124, 25 today, with 50 analysts. The analysts represent an apprenticeship model, so they. The long term, the medium term goal is as we deliver on the unit economics and the technology investment that Lisa outlined, that a lot of the stuff that the analysts do will no longer be done digitally.
The plan is that those analysts become the next generation of bankers, schooled in the Judo way of doing SME banking. I can. You know, we always run a list of people that we'd like to hire across each of our geographies. I can speak to Steve about the list that he's got of people that when we're ready to hire, who is it we would approach in the New South Wales market? When we're ready to hire in the Vic market, I can speak to Ben, and they'll produce a list of a dozen plus people that we have been talking to, we've got to know, they're keen to join, and we just keep that pipeline of future talent warm.
In fact, I'm having a catch up with one such person this evening, who's, you know, we've been talking to for a year, and he's really keen to join. It's just, it's not the right timing for us right now. I would say in six months time, we'll be ready to take him aboard. That talent management thing has been a big part of the way that we've thought about our business. You know, we don't wait to need to fill a vacancy. We make sure there's a pipeline and that the individuals come to Judo very warm in their knowledge of who we are and what we're trying to do, and that we're comfortable that culturally that they're a good fit, technically that they're a good fit, and they've got to pass a test. We could hire 50 bankers in the next two months.
From a pipeline of well over 100. It's not a problem for us because the proposition resonates with people that are passionate about SME banking and that are fed up with the complexity tax, the bureaucracy, and the slow-moving nature of others in the market. It's not a problem. It's just a question of managing, you know, how we do that.
Thanks, Joseph.
We've not lost anybody, actually. I should say. Oh, we lost someone. Sorry. Not to a major, a few weeks ago. We've lost about half a dozen bankers over the last three years. Four have gone to set up their own broker business. One's gone overseas on a holiday, and one has joined another bank, but not a major bank. That's the other thing we watch, because obviously when bringing in good people, you wanna keep them inside the company. That, for a range of reasons, not everybody stays forever. Having lost a small number of people, in a, in, you know, given the amount of hiring we've done, I think that's a big endorsement about the, how strongly our people feel about our company.
Cheers.
Thank you. O ur next question comes from the line of Josh Freiman from Macquarie. Please go ahead.
Hey, guys. Thanks for the opportunity to ask a couple of questions. Just two from me. The first is in relation to the chart on slide 19, where you guys give the indicative profile of investment spend. I mean, I'm conscious that you guys have a simpler bank, you know, less systems, less legacy tech, less issues. The reg and compliance spend looks particularly small, even relative to what I've seen from other sort of style of banks. I guess I just wanted to check, you know, how did you guys base your allocation for investment spend there? Because it feels like that portion is only gonna grow and grow from here.
Good morning. Thank you for the question. Yeah, because of the simplicity of the Judo model, and our operations, and that we have a simple set of products, the requirements around regulatory and investments are smaller compared to others. As I mentioned, we do not have any large remediation programs to date. That also makes a difference into the overall spend. What we try to do is effectively work on the things that matter, and there are things that will continue to grow. Like I mentioned, information security, that's one area that is really important that we continue to invest in. Then, from a compliance regulatory perspective, we continue to update things like, for example, open banking was the latest one for us.
You know, we delivered that in March, but that was a fraction of the spend of the major banks because of our simpler, simple tech stack, simple capabilities that we have here. It is based on two things. One is the current spend to date, and two, what we forecast the needs being from a regulatory environment and where we're taking our business. That's how we've done it, is basically looking at the past, but also forecasting where we think we'll need investments as we grow. For example, that's AML and fraud, et cetera. That's included in what we think we'll need to spend in the future.
Understood. Thank you. If I sort of reflect on my second question, if I look at your at scale targets, you have that CTI approaching 30. When I sort of consider that with the AUD 15 billion-AUD 20 billion, that's a pretty big range, which can drive the cost base as well, assuming your margins are over 3%. I guess with that 30% in mind, now that you've, you know, had a couple of years almost post IPO, how do you guys really view that 2026 growth rate for volumes?
Yeah, I mean, a couple of things on that. I think the, you know, on the ultimate kind of CTR landing point, you know, it's obviously influenced by a number of factors. We're still kind of a couple of years away, you know, from where we'll get to, you know, from where we'll get, you know, when we'll get to that at scale point. You know, clearly kind of cost is something that is in our control. You know, costs will continue to grow, you know, but at a lower rate, I think, than what you've seen us grow at in, you know, in the last couple of years.
You know, I think cost growth will be probably around, call it half, you know, next year what you've kind of seen, for this year, just as we start to see that cost airplane leveling out. We'll be more in those, you know, single digit kind of cost growth, as I mentioned in my presentation, going, you know, going forward from there. That really reflects the fact that you've got to do a lot of that infrastructure build up front. You know, you need to have the corporate, you need to have the risk, elements in place before you scale. So that's, you know, kind of reflecting that dynamic. The bigger kind of impacts then really are around what happens on the revenue side.
You know, there's a lot of factors that are influencing that. Where all of those kind of land is still, you know, we're a couple of years out for being more specific. Where funding costs land, it's clearly quite volatile. As Joseph mentioned, you know, where that ultimately lands, I think is something that we're watching and will be a component. What our funding mix ends up at the scale point. We've obviously provided some guidance and some ranges around how we see the stack of our funding, at that point.
You know, and other kind of dynamics in terms of margins that will impact that final landing point. I think it's fair to say that, you know, inflation has been a factor, obviously, that's impacted everyone. That has impacted our absolute CTR levels and probably means that from a lending book point, when we are at scale, we're probably, you know, more towards the upper end of that range that we've given. There's still a lot of factors that are here to play through, and a couple of years to go. We're very kind of comfortable with that approaching 30% CTR target.
I think also in the context of some of that U.K. benchmarking, which we've provided some disclosure on, kind of provides some context, you know, to that and where we'll get to. I think the other really important point here, though, is that, as I said, the AUD 20 billion level or when we're at scale is not our endpoint. We have a very big TAM. We will continue to grow. We won't be the biggest, as Joseph said, but we wanna be the best. The market opportunity is very much there, and we will continue to see those incremental benefits of ongoing scale above that point and we'll see the CTI on a downwards trajectory. I think that's also an important point to highlight.
Thank you.
Thank you. O ur next question comes from the line of Azib Khan from E&P. Please go ahead.
Thank you very much. Another question on margin. I know you've had a few on this, but another one from me. Just wanted to understand the increase in the trajectory of the margin on new TDs better. I think, Andrew, you said that the margin on new TDs in the March quarter was about 60 basis points. You told us back in February that the margin on new TDs in the month of January was 56. Can I take that to mean that the margin on new TDs in the month of March was about 64, 65 basis points? Is that about right? Then all of a sudden there's a step up to closer to 90 in April. Is that what's happened?
Yeah, no, thanks for the question, Azib. I mean, in terms of the, in terms of the margin that we've seen, what we have called out is the, is the 68 basis points, which is what we saw for the quarter. Look, the reality is that. On new TDs, I should state. We have seen the pricing kind of bounce around a little bit. I mean, we saw SVB, we saw the disruption that kind of had in terms of, in terms of markets. We saw offshore, markets for the big banks close in terms of, in terms of their ability to access. That, some of those factors have come back. The pricing has moved around, I think, during the quarter.
We did see it increase kind of post those events offshore, as people kind of stepped in to the domestic market to secure their funding needs. We saw it also move up, I think, in the first part of April, and it's come back a little bit over the last two weeks. We've seen a little bit of normalization there. The real driver, though, I think, of those movements has been probably more the swap movements than the absolute pricing. Certainly from kind of our perspective, we've been fairly consistent with our kind of leading pricing on our key tenors. You know, where Despite all of that kind of volatility and movement, we are now in the 80-90 basis point range. We think it'll be there for a period.
That's underscoring our underlying NIM guidance for the second half, and is, you know, based on history, where you need to be to be number one, and that's, you know, also baked into our at scale, NIM level. Things have moved around. There's been a bit of competition. It's been 68 basis points for the quarter. It's moved around a little bit, but it is now settled at that 80-90 basis point range, and it feels like it has settled a little bit there if we look at the last two weeks. That's our assumption, you know, for the rest of this financial year.
Andrew, with that last two weeks, yeah, I understand how you're saying it's come back a bit and the factors that you've mentioned that rose the increase. The last two weeks, is it fair to say then it's closer to 80 than 90?
It's probably in the upper part of that range. It moves around a bit. It moves depending on, you know, what the swap curve is doing. It moves depending on where we get our hedges in place as well, 'cause we look at all of this on a net hedge basis. You know, It's in that 80 to 90 basis point range. We're kind of comfortable with that, taking that as, you know, where it will be for the rest of this financial year. Certainly again, you know, looking at the historical levels that you'd need to be at, we're kind of comfortable with that as the assumption underpinning our at-scale NIM.
I understand that you're comfortable with that, with the rest of this year. Once we're into the September quarter, Andrew, can you see that cost of TDs rising above 90 at all?
Look, I mean, take some context, I guess, to what we've seen in March. Things were pretty disruptive. We were right in the thick of it. We were in the, we were close to blackout for a lot of the banks, so they couldn't do any other, any other funding. We saw all the disruption of what happened offshore. All of that noise, despite all of that noise, we've been in the 80 to 90 basis point range. Does it have the ability to go above that? Potentially. I think I take some comfort in what's happened in the last couple of weeks where that's settled, the swap curve's settled. It's, it's, it's dynamic. We're comfortable with that range, as I said earlier.
Just one more question, if that's okay. If I think about the NIM at the headline level and profit before tax, has there been any benefit to headline NIM and profit before tax in the March quarter from a rundown in liquid assets?
No, it's a, it's a great question, actually. Look, we're not, we're not dipping into liquids to manage NIM. Our MLH levels have been in that low 20% range. There's been no change in that stance over the, you know, from where we were at the first half. We're not gonna compromise on liquidity in this environment. You know, we haven't had to dip into that to prop up the NIM. The levels that we're, you know, we are in this environment, obviously kind of managing to a high level of liquidity, which we think is appropriate. There's been kind of no change in stance on liquidity for the quarter.
Just the final confirmation, Andrew. The increase in the underlying NIM guidance is purely from the marginal cost of new TDs being lower than expected. Is there anything else at all that's driving the increase in that underlying NIM guidance range?
Good question, because it is, TDs is part of the story, but the other component that has impacted that NIM is really the, some of our treasury activities. We've looked at the timing and, you know, we've optimized some of the management of those treasury activities. That's all about timing. There's no, kind of step change or change in strategy. It's just been about timing of some of those things. That's things like where we put the ultimate kind of hedging on, where we did the actual drawdowns for some of the new warehouses, how we seeded them. It's just been about optimizing those activities. It's been TDs, but it's also been the optimization of the treasury activities.
That's an important point because I think as we now look forward to the TFF refi, we look at what's ahead of us in terms of next year, optimizing that pathway is gonna be a real focus of management. It's something we are very focused on, because I think if you get that stuff right, it can have a benefit to NIM. We're gonna be very much kind of balancing that TFF refi against our funding and our liquidity and obviously our kind of, our NIM as well.
Great, Andrew. Just one more on that. Is it fair to say half of... You've shifted up your guidance range for the second half by 20 basis points. Would half of that be treasury activities?
We haven't kind of split that out. Look, both of them are important factors. I think the real point on this is this is just timing. Kind of where we've landed is where we've landed. You know, as we noted earlier with the Q&A, there's gonna be a step down from that level as the TD book kind of washes through, and ultimately as we start to repay the TFF. All of these things we kind of manage within the context of, you know, the context of the overall kind of treasury and funding management.
Thank you.
Thank you. I see our last question comes from the line of Richard Wiles from Morgan Stanley. Please go ahead.
Yeah, good morning, everyone. Lisa, I have a question about the cost of origination and servicing the loan. You mentioned it takes 60 hours to originate, 30 hours per year to service a loan. What's the AUD cost of that? What does the loan size and the loan duration need to be for you to deliver your low- to mid-teens ROE objective if you can keep the margin above 3%?
Very good question. We've used the average loan as the basis for the foundation of the numbers you saw today. Going forward, we're assuming the same portfolio, of loans is what we can project going forward, that it gives us confidence to achieve the CTI. We haven't disclosed today, what the actual costs are, and the reasons for that is it's quite a complex analysis and it's also marginal economics. It gets confusing with our broad details around how you could use it, you know, for in the instance you were trying to model. We have...
What's important about unit economics is actually the relative movement over time, which is why I stated that the unit economics as a measure, measures our scale each year, and in fact we'll do it twice yearly, our unit economics analysis. Over time, what we will measure is the relative improvements, like I showed in settlements and servicing today, and that will continue to go down. The reality is we have a lot of low-hanging fruit, and that's what you saw a sizable uplift today. I mean, the fact that we built our statement automation capability, you know, drove a 23% impact, you know, in six months, is really what's important and how to use the measure appropriately. Over time, we'll see those things go down as well as across the application and origination side as well.
Okay. Maybe if I could ask the question a different way. Do you need to grow the average size of loans in order to achieve your low- to mid-teens ROE objective if you can keep the 3% margin and if your unit economics go down over time?
No, we don't. Our analysis, it does not actually rely on growing or changing the mix of the loan book in any way. It doesn't assume that we could grow, have a larger proportion of larger loans and that would reduce, right, our unit economics. There is no assumption around changes in mix of the portfolio as where we land today. It's important.
Is the business old enough at this point to have a view on average loan duration? Have you seen, maybe Joseph, have you seen much rundown or turnover at the loan portfolio?
No. Well, from experience, the average loan duration on SME loan is 2.83 years. Whilst it's still a bit too early for us, we forecasted a 20% amortization in our loan book per annum. That 20% is made up, and is made up of about 10%-12% of principal repayment in the ordinary course. 5%-7% is actually refinancing out to another bank, at a lower Where it has happened, it's been at a lower interest rate. Then, you know, 2% or 3% has been one-off repayments from a business that may have sold a property or sold some other asset, has surplus cash and wants to pay down.
We monitor that like a hawk, the 20% assumption has held true. It's in a range of 17%-20%. You know, the big thing that we do focus in on is refinancing out to other banks. The issue there, I mean, that's been in the 5%-7%. It's held constant. It's largely, as I say, being where a business, for whatever reason, has chosen to refinance at a lower interest rate than we feel is acceptable given the risk profile of the business. That's a stat we look at every month in the business. It hasn't really changed over the course of the last 18 months or so.
Could I just ask one final question as well please, Joseph? What are you seeing on the front line when it comes to businesses that are exposed to the household sector and consumer spending? How's it tracking? How's their state of affairs tracking relative to your expectations across the broader economy?
Yeah. Well, I mean, when you look at the industry stats, and I know the insolvency stats that came out to the end of March showed that across the economy, that there had been an uptick in insolvency. I think in the month of March, it was 831 firms put into an administration. That compared to an average of 720 for the same month in the period 2016 to 2019, so pre-COVID. We have seen an uptick across the economy. Now, of the firms that have been put into administration in the nine months to the end of March, 1,600 have come from the construction sector, which has basically been about half of the total.
800 have come from accommodation and food services, 370 odd from retail services, and 342 from manufacturing. Our book is not... I mean, that's, those are the macro figures, which is signaling a definite uptick in businesses going into administration. Across our book is, I think as Andrew mentioned, we're not seeing any trends of retail versus accommodation leisure. We have very little in the way of direct construction risk. We've got some exposure, it's largely asset finance, it's equipment-based. Where there is a direct construction exposure, it's also secured against another underlying business. It's a business might want to build an extension to its premises, we'll finance that, it's secured against other cash flows.
We have little exposure to that direct exposure to that sector. I think those sectors that, you know, we are seeing a softening in manufacturing. We are seeing a continued softening in discretionary retail, as you would expect. A bit of that in hospitality as well. We'd expect that softening to continue as the real impact of the mortgage rate increase start to bite, which we've not seen yet. That'll really start happening over the course of the next six months. The book is, I think as Andrew mentioned, there's no emerging sector issue for us. The numbers still remain quite low, but we are very conscious that in a macro sense, there's evidence of certain sectors starting to show distress as the numbers that I just outlined indicated.
Thank you very much.
Thank you. That concludes our Q&A session. I would now like to turn the call back to Joseph Healy, CEO and Co-Founder, for closing remarks.
Thank you. Thanks to everybody that joined us this morning. I hope that you'll find the presentation to be information rich. I mean, that's always been our goal, as I mentioned at the beginning, that we want to be right up there in a market leading sense with our disclosures and the information around our business. We are conscious that a lot of people in the market are still learning and understanding the Judo business model, even though it's a very simple business model. The more information that we can provide the market to help inform judgment, that's something that is a very important goal in the way that we think about the investment community. Thank you for the questions because I felt there were some excellent questions this morning, and it's been very much appreciated.
I hope you leave this session with a much better knowledge about how we're gonna get to our, and how confident we are in getting to our low 30% CTI. The decisions that we're making and the investment in particular that we're making in leasing to make sure that we're building a world-class SME bank. I hope you also have left the session this morning feeling that Judo is probably one of the most undervalued stocks in the sector and has a screaming buy to be marked against it. Thank you for your participation and your time on Monday morning. Thank you very much.