Spline Data, a quantitative municipal bond data provider, has launched a product to give the muni market data-driven, actionable yield curves.
The company’s investment-grade, fixed coupon municipal bond curves aim to use every trade, in contrast to the current municipal market pricing tools, which often reflect only round lots— about 5% of trades — according to Spline Data founder and CEO Matthew Smith.
The muni market relies “upon noisy and fractured pricing data that is increasingly out of sync with today’s electronic trading ecosystems,” he said.
“With nearly 95% of trades in this market printing at $1 million in value or less, the need for quantitatively computed, independent municipal bond curves is clear,” Smith said.
Cloud computing allows the company to create more granular curves given the data, according to Smith.
“The appetite for purely quantitative market data is there,” he said. “We’re in a good position to grow in a direction that’s suited for customers’ needs and where the market is evolving.”
Smith wants to “eliminate the guesswork needed to execute in the municipal bond market by providing precise and frequent quantitative market data similar to what’s available in more mature asset classes.”
The curves are “purely data-driven, provable and transparent, providing the granular inputs that are vital to professional traders and asset managers in their algorithms, executions and analysis,” Smith said.
Spline’s curves are offered in a variety of configurations, including by rating, source of payback, and trade size, and they update every five minutes.
An important difference to other curves, Smith said, is how Spline frames the question.
“We’re asking if a trade were to happen in this specific subset of the market right now, say AAA, what yield would it most likely occur at,” he said.
“By framing it this way, we aren’t not beholden of waiting for that AAA trade to happen,” he said. “We can go back in time and look at hundreds of thousands of trades that have happened in the past, pass them through an array of models and make a well-informed estimate as to where they could indicate the AAA could happen.”
From there, Spline can use those metrics of “how well it answered the question” to improve its modeling.
This, he said, can then be deployed in a scalable environment to come up with estimates for hundreds of thousands of curves.